Elsevier

Neurobiology of Aging

Volume 22, Issue 4, July–August 2001, Pages 603-611
Neurobiology of Aging

Sex differences in corpus callosum size: relationship to age and intracranial size

https://doi.org/10.1016/S0197-4580(01)00232-9Get rights and content

Abstract

This quantitative MRI study reports measurement of corpus callosum area taken from midsagittal brain images in 51 healthy men and 41 healthy women, spanning the adult age range (22 to 71 years). Men had larger brains and corpora callosa than women, but callosal size did not correlate with age in either sex. Intracranial (i.c.) volume (ICV) and midsagittal i.c. area (ICA) of brain were used in covariate, regression, and ratio analyses to determine whether sex differences in the corpus callosum endured with statistical adjustment for sex differences in maximally attained brain size. With the exception of one ratio measure, the different statistical adjustments for the contribution of sex differences in brain size to corpus callosum size all indicated that men had larger corpora callosa than women for their brain size. A subsample of men and women selected to be matched on i.c. volume and age confirmed this statistical observation. Sexual dimorphism in the corpus callosum is not a simple artifact of sex differences in brain size and may reflect differences in connectivity necessitated by differences in brain size.

Introduction

The corpus callosum, a prominent body of white matter readily visualized with magnetic resonance imaging (MRI), is the thick band of white matter fibers that connects the two cerebral hemispheres in a homotopic organization with reference to the cortex [31]. The corpus callosum is of interest not only because of its key role in normative processes of hemispheric communication and specialization [42] but also because of its vulnerability to environmental toxins, including alcohol [19], [24], [33], and other white matter diseases such as multiple sclerosis [3]. Normative MRI studies of the corpus callosum in men and women spanning the adult age range typically use a single midsagittal slice from which to measure area or shape (for reviews, see [16], [41]) and have employed a variety of approaches for defining the callosal subregions. Histologic studies have shown that white matter tracts occupying anterior (genu + rostrum), medial (body), and posterior (splenium) regions of the corpus callosum primarily link different anterior to posterior cortical association areas in each hemisphere [10], [26].

Accurate measurement of the size of this structure and its subregions in healthy men and women across the adult age range can estimate the normative aging process and sex differences and provide age and sex norms against which to compare patient groups. Current results indicate that brain white matter volume, including the corpus callosum, expands greatly during childhood continuing into late adolescence and possibly beyond [1], [46], and its growth is singularly critical in determining maximum i.c. volume (ICV) [34]. Once maturation is complete, however, cortical white matter volume remains relatively stable, at least through the seventh decade [34], [36]. Most cross-sectional MRI studies of the corpus callosum likewise show this structure to be relatively immune to age-related shrinkage in adults from the third to seventh decade of life [4], [7], [16], [22], [33], [41]. By contrast, some studies have shown senescent effects over the third to eighth decade [15], [43], and others have found aging effects in elderly samples, especially those exceeding age 55 years (e.g. [9], [37]). Weis et al. [43] reported aging effects in younger decades, but this group comprised headache clinic patients with negative MRI clinical report. Of these MRI studies, some have shown greater vulnerability to aging in anterior than posterior regions of the corpus callosum (e.g. [37], [43]), particularly marked in women (e.g. 37), while others find greater vulnerability of men compared to women in the older decades (e.g. 22).

Following an autopsy report that women have larger and more bulbous splenia than men [11], some subsequent MRI studies have supported this sex difference in shape [2], [9], [37]. Other studies reported equivalent [2], [5], [6], [23], [44], [45] or larger [21], [22], [37] sized corpora callosa in women than men once brain-size differences were taken into account. To address these discrepancies, meta-analysis [16] including both pathological and MRI studies (n = 36) found a mean effect size of 0.27 for raw corpus callosum area (men > women). In the subset of studies (n = 11) that expressed corpus callosum area as a ratio of midsagittal brain area, however, the effect size was negative (–0.26) and the authors concluded that callosal area adjusted for brain size was larger in women than men [16].

Men generally have larger bodies and brains than women; therefore, brain size must be accounted for in sex comparisons of the corpus callosum. Because the relationship between corpus callosum size and brain size is not necessarily linear and may itself vary between men and women, inconsistent reports on sex differences in the size of the corpus callosum and its subregions may reflect the different approaches taken, not only to measure but also to account for brain size. For example, some measures of brain size are based only on brain tissue (e.g. 21) even though age-related decline in brain tissue volume from its maximally attained size limits the value of using tissue alone to correct for sex differences in gross brain size.

In this study, we examined age effects and sex differences in corpus callosal size across the adult age range in a sample of healthy community dwelling men and women. In addition to analyzing uncorrected (raw) measures of corpus callosum, we used two estimates of maximally attained brain size, i.c. volume (ICV) and midsagittal i.c. area (ICA), and three statistical approaches to account for sex-related differences in these measures of maximally attained brain size: covariation, regression, and ratios. The covariation approach entered callosal values of both men and women into an analysis of covariance with ICV or midsagittal ICA as a covariate. The regression approach regressed corpus callosum measures on ICV or midsagittal ICA for the entire study sample to derive residuals that would express the extent to which each individual’s corpus callosum deviated in size from mixed gender population norms for someone of similar brain size [28], [29]. The ratio approach expressed each subject’s corpus callosum measures as a ratio of his or her own ICV or midsagittal ICA. Covariation and residualized approaches are influenced by variations in ICV or midsagittal ICA across the entire sample (both men and women) and are dependent on all subjects sampled, whereas the ratio approach assumes a simple linear relationship between corpus callosum and brain size across all subjects and expresses each individual’s corpus callosum measure relative to his or her own ICV or midsagittal ICA. Thus we expected different outcomes for each approach,

We predicted that age would not affect corpus callosal size over the age range in this sample because brain development was completed [12] and because a prior study of a similarly recruited sample detected no age cohort effects in cortical white matter volumes [34]. By contrast, we expected men to have larger callosal measures without statistical adjustment for differences in maximally attained brain size. Although we expected some difference in outcomes for population-based versus individual approaches to head-size correction, we expected that sex differences would be significantly attenuated or entirely removed with adjustment for brain size regardless of the brain size adjustment approach applied.

Section snippets

Subjects

The subjects included 51 men (mean age ± SD = 45.2 ± 13.9, range = 23 to 71 years) and 41 women (mean age ± SD = 49.1 ± 16.6, range = 22 to 71 years), recruited from the local community. The men were described in an earlier report [33]. Subjects who passed a telephone screen were interviewed by a psychiatrist or psychiatric research fellow using the Schedule for Affective Disorders and Schizophrenia (SADS)-lifetime [18] or the Structured Clinical Interview for Diagnosis (SCID) [39] as

Age and sex effects on corpus callosum and brain size

Raw total corpus callosal area was not significantly correlated with age in either men or women (Fig. 2). The lack of correlation with age also held for the two residualized measures and the two ratio measures. Furthermore, neither ICV nor midsagittal ICA correlated significantly with age in either men (ICV r = −0.003; ICA r = −0.20) or women (ICV r = .10; ICA r = −0.12)

Similarly, raw regional corpus callosal areas were not associated with age: splenium (men r = .05, women r = .14), body (men

Discussion

This quantitative MRI study provided evidence for significant sexual dimorphism in total cross-sectional areas of the corpus callosum. Because of the strong relationship between callosal size and overall brain size, which also shows sexual dimorphism, we employed several statistical approaches to account for these differences. Use of variation in i.c. size—whether a volume or an area estimate—across the combined sample of men and women (covariation or calculation of standardized residuals) to

References (46)

  • T. Passe et al.

    Age and sex effects on brain morphology

    Progress In Neuro-Psychopharmacology & Biological Psychiatry

    (1997)
  • A. Pfefferbaum et al.

    Brain structure in men remains highly heritable in the seventh and eighth decades of life

    Neurobiol Aging

    (2000)
  • D. Salat et al.

    Sex differences in the corpus callosum with aging

    Neurobiol Aging

    (1997)
  • L.S. Allen et al.

    Sex differences in the corpus callosum of the living human being

    J Neurosci

    (1991)
  • F.J. Barkhof et al.

    Functional correlates of callosal atrophy in relapsing-remitting multiple sclerosis patients. A preliminary MRI study

    J Neurology

    (1998)
  • W. Byne et al.

    Variations in human corpus callosum do not predict gendera study using magnetic resonance imaging

    Behav Neurosci

    (1988)
  • C. Cowell et al.

    A developmental study of sex by age interactions in human corpus callosum

    Developmental Brain Research

    (1992)
  • H.F. Crovitz et al.

    Group test for assessing hand and eye dominance

    Am J Psychol

    (1962)
  • C. Davatzikos et al.

    Sex differences in anatomic measures of interhemispheric connectivitycorrelations with cognition in women but not men

    Cerebral Cortex

    (1998)
  • M. de Lacoste et al.

    Topography of the human corpus callosum

    J Neuropath Exper Neurol

    (1985)
  • A. Dekaban et al.

    Changes in brain weights during the span of human liferelation of brain weights to body heights and body weights

    Ann Neurol

    (1978)
  • A. Deshmukh et al.

    Quantification of cerebellar structures with MRI

    Psychiatry Res: Neuroimaging

    (1997)
  • J. Desmond et al.

    On-, and off-line Talairach registration for structural, and functional MRI studies

    Human Brain Mapping

    (1997)
  • Cited by (0)

    Support for this study was provided by AG17919, AA10723, AA05965, AA12388, MH58007, and the Department of Veterans Affairs. A partial report of these data was presented at the annual meeting of the International Neuropsychological Society, February 1999 [40].

    View full text