|Home | About | Journals | Submit | Contact Us | Français|
The human brain uncinate fasciculus (UF) is an important cortico-cortical white matter pathway that directly connects the frontal and temporal lobes, although there is a lack of conclusive support for its exact functional role. Using diffusion tensor tractography, we extracted the UF, calculated its volume and normalized it with respect to each subject’s intracranial volume (ICV) and analyzed its corresponding DTI metrics bilaterally on a cohort of 108 right-handed children and adults aged 7–68 years. Results showed inverted U-shaped curves for fractional anisotropy (FA) with advancing age and U-shaped curves for radial and axial diffusivities reflecting white matter progressive and regressive myelination and coherence dynamics that continue into young adulthood. The mean FA values of the UF were significantly larger on the left side in children (p=0.05), adults (p=0.0012) and the entire sample (p=0.0002). The FA leftward asymmetry (Left > Right) is shown to be due to increased leftward asymmetry in the axial diffusivity (p<0.0001) and a lack of asymmetry (p>0.23) for the radial diffusivity. This is the first study to provide baseline normative macro and microstructural age trajectories of the human UF across the lifespan. Results of this study may lend themselves to better understanding of UF role in future behavioral and clinical studies.
The human uncinate fasciculus (small hook-shaped bundle) connects the inferior frontal gyrus and the inferior surfaces of the frontal lobe with the anterior portions of the temporal lobe (Ebeling and Cramon, 1992; Kier et al., 2004; Schmahmann and Pandya, 2006). The exact role of the UF’s function and its relevant human behavioral relations are not yet resolved (Catani and Mesulam, 2008; Duffau et al., 2009; Parker et al., 2005). The UF is traditionally considered to be part of the limbic system and is known for its involvment in human emotion processing, memory and language functions (Schmahmann et al., 2008).
To the best of our knowledge, Highley et al. (2002) is the only postmortem histological study on the human UF that reported a rightward asymmetry of the area and fiber density for this structure in older adults. The UF has been used as a marker of tissue integrity in noninvasive magnetic resonance imaging (MRI) studies in healthy (Catani et al., 2002; Mamata et al., 2002) and diseased (Antunes et al., 2002; Aralasmak et al., 2006; Borroni et al., 2008; Kier et al., 2004; Kezele et al., 2008; Hasan et al., 2009b) populations. Other DTI reports on the UF includes using region-of-interest (Kubicki et al., 2002), voxel-based (Burns. et al., 2003; Park et al., 2004), and fiber tracking deterministic approaches (Catani et al., 2002; Mori et al., 2002) and probabilistic methods (Hagmann et al., 2003) to clarify changes related to tissue maturation (Dubois et al., 2006), development (Constable et al., 2008; Eluvathingal et al., 2007; Lebel et al., 2008) and normal aging (Zahr et al., 2009). The UF has also been studied using DTI methods in several clinical conditions that include frontotemporal dementia (Fujie et al., 2008; Matsu et al., 2008), Alzheimer’s (Taoka et al., 2006), mental retardation (Yu et al., 2008), social deprivation (Eluvathingal et al., 2006), depression (Sheline et al., 2008), bipolar disorder (Houenou et al., 2008), schizophrenia (Nakamura et al., 2005; Kanaan et al., 2005; Price et al., 2008; Szeszko et al., 2008), spina bifida (Hasan et al., 2008), traumatic brain injury (Niogi et al., 2008), temporal lobe epilepsy (Diehl et al., 2008; Rodrigo et al., 2007; Mcdonald et al., 2008) and multiple sclerosis as a marker of Wallerian degeneration (Hasan et al., 2009b).
In general, there is limited literature on the normal aging trajectories and normative values for UF macro (morphometry) and micro-structures (morphology) (Schmahmann and Pandya, 2006). Age has been shown to be a confounding factor in the interpretation of clinical data from structures such as the UF (Jones et al., 2006). Due to differences in acquisition and analysis methods between studies with relatively small population size, the available DTI reports (see Table 1) have provided a conflicting and incomplete account of the UF age, gender and diffusion anisotropy asymmetry in both children and adults. To our knowledge, there has been no cross-sectional quantitative MRI study on healthy controls across the lifespan for evaluation of the UF macrostructural (e.g. volume) and microstructure (e.g. diffusion indices).
In this work, we applied DTI fiber tracking methods as described previously (Eluvathingal et al., 2007) on a cohort of right-handed children and adults to quantify the spatiotemporal trajectories of the UF volumes bilaterally along with its corresponding DTI indices such as fractional anisotropy, radial and axial diffusivities. Given previous DTI works on whole brain white matter (Hasan et al., 2007b) and regional white matter tracking trends (Hasan et al., 2009a) that identified nonlinear relationships of DTI metrics with age, we hypothesized that the microstructural attributes of the UF development and aging trajectories would be best characterized by nonlinear curves across the human lifespan.
Group mean values and comparisons of the intracranial volume (ICV) and the right and left UF volume (UFV) on all possible subgroups of study participants (boys, girls, men, women, children and adults) are summarized in Table 2. The ICV was not significantly different between boys and men (p = 0.27). The ICV was not significantly different between girls and women (p=0.08). The ICV was significantly larger in males compared to females (p<0.000001; see Table 2). The ICV and right UFV were larger in males compared to females (p<0.005). The UF volume-to-ICV ratios (UFV/ICV × 100%) were not significantly different between males and females (p>0.14).
The UF corresponding FA, radial and axial diffusivity are summarized in Table 3. There were no significant differences between males and females (p>0.14; see Table 3). Therefore, age effect results are from a pooled sample of males and females.
Pearson correlation coefficients of the UFV/ICV, FA, radial and axial diffusivities and age for all children and adults are summarized in Table 4. Note the lack of dependence on age in the normalized UFV between children (p>0.2) and adults (p>0.13). The DTI metrics in Table 4 show that children have significant linear trends compared to adults which necessitated quadratic least-squares modeling for the age effects. Table 5 summarizes the least-squares coefficients using the quadratic least-squares (QLS) model, standard deviation and significance. The scatter data, least and quadratic least-squares fits of the normalized UF volume, corresponding FA, radial and axial diffusivity are plotted in Figure 1 and Figure 2 for both the right and left UF respectively.
There were no statistically significant age-by-sex or age-by-hemisphere interactions. Comparison of linear and quadratic coefficients for both males and females showed no differences between males and females (p>0.3). The absolute and ICV-normalized UF volumes for the right side were larger as compared with the left in children (p=0.02) but not significantly different either in adults (p>0.56) or in the entire population (p=0.27). The mean FA values of the UF were significantly larger on the left side in children (p=0.05), adults (p=0.0012) and the entire sample (p=0.0002). This leftward asymmetry (Left > Right) in the diffusion anisotropy is explained by a leftward asymmetry in the axial diffusivity (p<0.0001) and a lack of asymmetry (p>0.23) in the radial diffusivity.
The UF has been used as a marker of development, natural aging and disconnection of the temporal and frontal lobes that result from surgical interventions (Kier et al., 2004; Wurm et al., 2000; Schoene-Bake et al., 2009), congenital (Hasan et al., 2008), traumatic (Levine et al., 1998), and degenerative disorders (Aralasmak et al., 2006; Kanaan et al., 2005; Hasan et al., 2009b). Quantitative noninvasive MRI literature on the UF across the human lifespan is generally scant. To the best of our knowledge, this is the first cross-sectional lifespan study on the development and aging of the UF in both hemispheres using DTI fiber tracking methods. We have presented and compared both absolute UF volumes and ICV-normalized values to reduce effects related to human brain size variability (Im et al. 2008; Walhovd et al., 2005).
The ICV values in our healthy cohort are comparable to those reported in other studies in controls (Buckner et al., 2004). The fiber-tracked UF volume and its ICV-normalized values exhibited larger variability compared with the corresponding DTI metrics. Large variability in fiber tract volume using DTI has also been reported on the cingulum (Heiervang et al., 2007), the corticospinal tract (Reich et al., 2006), and uncinate fasciculus (Wakana et al., 2007). To the best of our knowledge, only three DTI publications documented the UF volume and its corresponding DTI metrics on healthy adults (Malykhin et al., 2009; Taoka et al., 2006; Wakana et al., 2007; see Table 1). Some studies on the UF using coronal 2D ROIs may have included the inferior fronto-occipital fasciculus (Catani et al., 2002; see Kanaan et al., 2005 for details).
The volume measurements of the UF in our adult population (see Table 2) are larger than those reported previously (Malykhin et al., 2009; Taoka et al., 2006; Wakana et al., 2007) in adult controls (see Table 1). These differences may be attributed to differences between samples used (e.g. age, brain size), DTI acquisition paradigm (e.g. voxel size, signal-to-noise ratio) and analysis procedures (e.g. tracking thresholds). Consistent with previous reports in the healthy UF (see Table 1), we report no sex differences on the normalized UF and its corresponding DTI metrics.
Our DTI results on the development and aging of the UF consolidate previous normative studies (see Table 1) that reported positive linear age trends in healthy children (Eluvathingal et al., 2007) and young adults (Lebel et al., 2008), and negative age trends in older adults (Jones et al., 2006). The current work UF DTI age trajectories resemble those published on whole brain white matter (Hasan et al., 2007b), lobar (Sowell et al., 2003) and subcortical volumes (e.g. hippocampus and amygdala see Walhovd et al., 2005).
The DTI-related metrics (FA, eigenvalues) provide complementary information about the microstructural substrates of tissue organization (Beaulieu, 2002; Hasan et al., 2009b). In particular, the decrease in the radial eigenvalues during childhood and increase during adulthood with advancing age may offer early predictors of the regional dynamics of myelination and demyelination (Song et al., 2005).
Our DTI results reproduce an inconsistently reported UF leftward asymmetry of the diffusion tensor anisotropy (Table 1) in healthy children (Eluvathingal et al., 2007) and adults (Kubicki et al., 2002). This trend has not been reported in other studies (Taoka et al., 2006; Lebel et al., 2008; see Table 1) or was even reversed (Rodrigo et al., 2006). A previous postmortem study on the UF cross-sectional area in older adults (Highley et al., 2002; see Table 1), however, reported a rightward asymmetry in UF fiber density, and no age or sex effects.
In our study, the UF diffusion tensor anisotropy leftward asymmetry is more specifically related to a statistically significant leftward asymmetry in the axial diffusivity which could result from a more coherent or less tortuous alignment of fibers (Beaulieu, 2002; Takahashi et al., 2000). Our lifespan cross-sectional study in healthy right-handed controls shows a dissociation between volume-based or macrostructural asymmetry and DTI-based or microstructural asymmetry. Our results on the axial asymmetry should be interpreted carefully as the UF right and left tract volumes were measured using diffusion tensor tractography which is limited by the DTI acquisition spatial resolution and the tracking procedures adopted. The biophysical substrates of this finding require histochemical and postmortem analyses (Beaulieu, 2002; Highley et al., 2002; Takahashi et al., 2000).
Our results provide useful baseline data on the effects of side, gender and age on the UF volume and its corresponding DTI metrics. Our results should help in the experimental design of future clinical investigation of the exact functional role of the U.
Access to anatomic information and the pattern of normal developmental trajectory of the uncinate fasciculus expands the breadth and depth of our knowledge of this structure (Schmahamn and Pandya, 2006). The significant role of these patterns of information may play an important role in evaluating fronto-temporal structural and functional connectivity in clinical or psychiatric populations. Since the UF originates from the anterior temporal lobe in front of the temporal horn and the cortical nuclei of amygdala and terminates in several critical areas of the frontal lobes such as orbital frontal cortex. As such the uncinate fasciculus is especially considered to be a critical structure in playing an important role in memory and emotion and our further understanding of these structures may offer new opportunities to provide clinically significant services to patient populations with specific neurologic or psychiatric dysfunctions. An example of such relationship has been recently shown by Fujie et al. (2008). Therefore, our current work on the normal structural development of uncinate fasciculus generates excitement for the expansion of our limited knowledge of brain fiber bundles and their mysterious functional significance.
Our normative database has been formed by pooling cross-sectional data collected on healthy children and adults using the same DTI protocol (study span ~ 3 years) to help in the interpretation of data collected from patients (Hasan et al., 2008; Hasan et al., 2009b). Our cross-sectional cohort and the lifespan experimental design accounted for the confounding and nonlinear age effects and warrant future longitudinal studies on the structural basis of functional disconnectivity.
Future extensions of the current studies include the investigation in larger samples of males and females of the interplay between UF volume, DTI metrics and the corresponding cortical gray matter thickness and subcortical substrate volume (e.g. hippocampus and amygdala) in both health (Sowell et al., 2003; Walhovd et al., 2005) and disease (Aralasmak et al., 2006; Highley et al., 2002; Kier et al., 2004).
This study included a total of 108 healthy right-handed children, adolescents and adults (age range 6.7–68.3 years). The cutoff age for children and adolescents was 20 years (Table 2). The boys/girls, men/women and males/females groups were age-matched (p>0.2). All participants were primarily English-speaking, identified as neurologically normal by review of medical history, and were healthy at the time of the assessments. The MRI scans were read as “normal” by a board certified radiologist (LAK). Written informed consent from the adults, guardians and adolescents, and assent from the children participating in these studies was obtained per institutional review board regulations for the protection of human subjects.
All MRI acquisitions were performed on a 3T Philips Intera scanner with a dual quasar gradient system with a maximum gradient amplitude of 8 Gauss/cm and an eight channel SENSE-compatible head coil(Philips Medical Systems, Best, Netherlands).
The DTI studies utilized a high signal-to-noise ratio (SNR) whole brain DTI protocol that was kept under 7 minutes (Hasan et al., 2007a). The DTI data were acquired using a single-shot spin-echo diffusion sensitized and fat-suppressed echo-planar imaging sequence with the balanced Icosa21 tensor encoding scheme (i.e. twenty-one uniformly-distributed directions over the unit hemisphere) (Hasan, 2007), b-factor = 1000 sec mm−2, repetition time (TR)/echo time TE = 7100/65 msec. The echo-planar phase encoding used a SENSE k-space undersampling factor of two (R=2) with an effective k-space matrix of 112×112, image matrix after zero-filling = 256×256. The field-of-view = 240 mm × 240 mm, number of axial sections = 44 with no gap and slice thickness = 3 mm. The number of b~0 magnitude image averages was 6; in addition, each diffusion encoding was repeated twice and magnitude-averaged to enhance the SNR (Hasan, 2007). The SNR, DTI data spatial resolution (e.g. voxel size 2.2 mm × 2.2 mm × 3 mm) and image quality (e.g. reduced image distortions) were adequate to trace the uncinate fasciculus (Eluvathingal et al. 2007; Fujiwara et al., 2008; Hasan et al., 2009a).
In this work, the DTI-derived rotationally-invariant metrics included the fractional anisotropy (FA), axial (L1 = λ||) and radial diffusivity (LT = λ). The radial diffusivity is defined as the average of the second and third eigenvalues (λ = (λ2+λ3)/2) and has been shown by several researchers to be a marker of myelination (Beaulieu, 2002; Song et al., 2005; see also Hasan, 2006). We performed distortion correction to remove eddy-current artifacts, then calculated the ICV from the non-diffusion weighted image (Hasan et al., 2009a). Additional details of DTI image processing (Hasan et al., 2007b; Hasan et al., 2009a) and DTI quality control measures are found elsewhere (Hasan, 2007a).
Compact white matter fiber tracking was performed using DTI Studio software (Jiang et al., 2006) based on the fiber assignment by continuous tracking (FACT) algorithm. Diffusion tensor tractography using anatomical landmarks (Kier et al., 2004) and multiple regions-of-interest (ROI) were utilized to trace the UF bilaterally as described elsewhere (Catani et al., 2002; Mori et al., 2002; Jones et al., 2006; Wakana et al., 2007; Yu et al., 2008). The UF tracking used a fractional anisotropy (FA) threshold of 0.15 and angle threshold of 60 degrees. These tracking thresholds were identical to those adopted to trace the UF in two previous publications using the same DTI protocol (Eluvathingal et al., 2007; Hasan et al., 2008). Additional details of the fiber tracking procedures are provided elsewhere (see APPENDIX in Eluvathingal et al., 2007).
In brief, the anterior commissure (AC) identified on a coronal DTI color-coded plane was used as an anatomical landmark to identify the two regions where the UF superior and inferior segments traverse the coronal plane (Figure 3A regions 1 and 2). Note that the seed ROIs selected (Figure 3A) border the external capsule, insula, putamen and the hippocampus (Ebeling and Cramon, 1992). A Boolean “OR” operation on region 1 (lower portion of external capsule) combined with an “AND” operation on region 2 (temporal lobe) were sufficient to construct the UF on each side avoiding any mixing with other tracts such as the inferior fronto-occipital and inferior longitudinal fascicule (Catani et al. 2002; Mori et al. 2002). Once the UF fiber tract is reconstructed, its volume and corresponding DTI attributes were recorded and visualized in 3D (Figure 3C and 3D).
All analyses of UF absolute, normalized volumes and their corresponding DTI metrics variation were conducted using a generalized linear model with effects of age, side and sex as described elsewhere (Hasan et al., 2007b). Given previous reports on white matter macro and microstructural development across the lifespan (Sowell et al., 2003; Hasan et al., 2007b; Hasan et al., 2009a), both linear and quadratic age terms were included. The DTI metrics (e.g., fractional anisotropy = FA; radial or transverse diffusivity = λ = LT; axial diffusivity = λ|| = L1) were modeled for both males and females as yf= β0+β1*age+β2*age2, and then the general least-squares methods were used to compute the coefficients, standard errors and their significance using analysis-of-variance methods (Hasan et al., 2009a). Interactions of sex with age (both terms) were examined, and trimmed where non-significant. If the quadratic age term was not significant, age and sex interactions were examined without this term, and trimmed if non-significant. All statistical analyses were conducted using SAS 9.1 (SAS Institute Inc, NC) and MATLAB R12.1 Statistical Toolbox v 3.0 (The Mathworks Inc, Natick, MA).
This work was funded by NIH-NINDS R01 NS052505-04 awarded to KMH, NIH-NICHD R01 NS046308 awarded to LEC, P01 HD35946 awarded to JMF and 1 P01 NS46588 awarded to ACP. The authors wish to thank Vipul Kumar Patel for helping in data acquisition.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.