Longitudinal study of carbon monoxide intoxication by diffusion tensor imaging with neurospsychiatric correlation
=================================================================================================================

* Chiung-Chih Chang
* Wen-Neng Chang
* Chun-Chung Lui
* Jiun-Jie Wang
* Chih-Feng Chen
* Yu-Chang Lee
* Shun-Sheng Chen
* Yu-Ting Lin
* Chi-Wei Huang
* Ching Chen

## Abstract

**Background:** White matter damage is common after carbon monoxide (CO) intoxication, but in vivo follow-up studies about the mechanism of white matter damage are not possible in pathology series. Diffusion tensor imaging (DTI) and voxel-based morphometry (VBM) can quantify diffusion parameters and volumetric changes in white matter that can be correlated with neuropsychological performances in longitudinal studies.

**Methods:** We examined 9 patients with CO intoxication using DTI, VBM and neuropsychologic tests at an average of 3 and 10 months after CO exposure. We used data from 18 age- and sex-matched controls for comparison.

**Results:** We found that cognitive recovery at 10 months after CO intoxication was not significant, although it was after 3 months. The neuropsychologic tests correlated better for the fibre tract of the semicentrum ovale and not the periventricular fibres. Diffusion measures suggest increases in fractional anisotropy, mean diffusivity and axial eigenvalues over time, while increases in radial eigenvalue were evident at 3 months compared with controls. Periventricular white matter atrophy was observed 10 months after CO intoxication

**Limitations:** Our study included few cases, and the interpretation of the putative changes on neuroimaging findings cannot be confirmed by histology.

**Conclusion:** Our study showed that the evolution of white matter injury in CO encephalopathy occurred over time. Cognitive recovery was not evident in the follow-up period because of white matter injuries.

## Introduction

Carbon monoxide (CO) intoxication by charcoal burning has become the primary method of attempted suicide in Taiwan,1 leading to an increase of more than 20% in the overall suicide rate.2 Carbon monoxide intoxication can lead to acute and chronic neuropsychiatric complications. The pathophysiologic mechanisms of brain injury following CO exposure vary.3 One important factor is the inflammatory reaction after toxic exposure.4 Cortical and subcortical white matter damage have been reported to account for the neuropsychologic deficits,5–7 but longitudinal follow-up of white matter damage is not possible in pathology series.

Voxel-based morphometry (VBM) is an unbiased, whole-brain quantitative method that can detect volume changes of grey8 and white matter. Diffusion tensor imaging (DTI)9 is a recently developed tool that uses magnetic resonance imaging (MRI) technology. Fractional anisotropy (FA) and mean diffusivity are 2 DTI-derived parameters that provide information about microstructure changes.10 Although the specifics are not well known, FA quantifies the directionality and coherence of the white matter tracts by factors such as myelination, axonal density and integrity.11,12 Analysis of axial and radial diffusivity may provide potential measures of the mechanisms of white matter changes because increased axial diffusivity has been reported to reflect axonal changes in trauma, whereas changes in radial diffusivity strongly correlate with myelin abnormalities.13 In this study, a combination of DTI and VBM were used to monitor CO-intoxicated patients over time. Their clinical parameters were compared with those of controls.

## Methods

### Patient enrolment

The neurology clinic at Kaohsiung Chang Gung Memorial Hospital recruited 9 CO-intoxicated patients from August 2006 to November 2008. We included 18 age-matched healthy participants from the normative database as controls for neuropsychologic testing and MRI comparison. The age of each control was within 2 years of the sex-matched CO patient. None of the control participants had a history of neurologic or neuropsychiatric disorders, and all had normal results of MRI scanning and basic blood tests (liver and renal function tests, electrolytes and complete blood cell counts).

Informed consent was obtained from all participants, and the Institutional Review Board of Chang Gung Memorial Hospital approved the study protocol.

### Cognitive testing

We used a comprehensive battery of tests to assess the cognitive ability of all participants and the Mini-Mental State Examination (MMSE)14 to assess their general intellectual function. We evaluated cognitive severity using the Clinical Dementia Rating (CDR) Scale.15 We also assessed verbal and nonverbal episodic memory using the word-recognition test from the Alzheimer’s Disease Assessment Scale–Cognitive (ADAS-Cog)16 and recollection of the Rey–Osterrieth complex figure17 after a 10-minute delay (maximal value = 17, normal value > 13.4). We chose CDR because it evaluates the functional capacity of the participant independent of physical disability, which may occur in patients with CO intoxication. The ADAS-Cog memory subset was chosen to investigate memory recall and registration ability because it has been extensively tested for its sensitivity and specificity in differentiating between patients with and without memory problems.

We adapted the test for memory-retrieval ability from the word-recognition test of ADAS-Cog. We evaluated the scores as follows: a hit (maximal value = 12, normal value > 9.5) was the recognition of a word that had appeared before; a false alarm was a response to a word that was not in the test list; a miss was a nonrecognition of a word that had appeared before; and a correct rejection (maximal value = 12, normal value > 8.5) was the recognition that a word had not appeared before. We used language screening tests including the 17-item Boston Naming Test18 (maximal value = 17, normal value > 16.5) and semantic verbal fluency (number of animals and vegetables in 1 min).

We assessed visuospatial ability by instructing participants to copy a modified Rey–Osterrieth complex figure (maximal value = 17, normal value > 16.3) and by use of the number–location test from the Visual Object and Space Perception Battery19 (maximal value = 10, normal value > 9.5). We also assessed the ability to perform 5 arithmetic calculations and digit forward span. We evaluated executive function using the digit backward span, reasoning and abstract thinking questions and design fluency test.20 We used the Neuropsychiatric Inventory (NPI) to assess behavioural symptoms.21

### Magnetic resonance imaging protocols

Scans were performed using a 3-T scanner (Excite; GE Medical System) equipped with echo-planar capability. The structural MRI sequences were as follows:

1.  Axial fast spin-echo *T**2*-weighted image: repetition time 4200 ms, echo time 102 ms, 2 acquisitions, field of view (FOV) 240 × 240 mm, 320 × 224 matrix, section thickness 5 mm.

2.  Axial fluid-attenuated inversion recovery image: repetition time 8000 ms, echo time 100 ms, time for inversion 2000 ms, 1 acquisition, FOV 240 mm × 240 mm, 320 × 256 matrix, section thickness 5 mm.

3.  *T*1 inversion recovery–prepared 3-dimensional spoiled gradient-recalled acquisition in steady-state sequence: repetition time 8600 ms, preparation time 400 ms, FOV 240 mm × 240 mm, slice thickness 1 mm.

4.  Diffusion tensor imaging: single-shot echo-planar sequence with gradients applied in 25 noncollinear directions. Axial images were acquired using the following parameters: repetition time 7000 ms, echo time 72 ms, FOV 240 mm × 240 mm, 128 × 128 matrix, which led to an in-plane resolution of 1.875 mm. Data were subsequently interpolated to 1 × 1 mm. Thirty contiguous slices of 5-mm thickness were obtained without gap. A b value of 1000 s/mm2 was used.

We processed the imaging data on a personal computer using SPM 5 ([www.fil.ion.ucl.ac.uk/spm/](http://www.fil.ion.ucl.ac.uk/spm/); Wellcome Department of Cognitive Neurology, London, UK) for VBM data and FSL version 4.0.1 ([www.fmrib.ox.ac.uk/fsl/](http://www.fmrib.ox.ac.uk/fsl/)) for DTI data.

### Voxel-based morphometry

The VBM protocol followed standard procedures.22 We created an ad hoc age-matched template image to optimize the spatial normalization. We applied affine and nonlinear transformations to spatially normalize the participants’ images to the template. We sequentially performed normalization, segmentation into 3 classes of tissue (grey matter, white matter and cerebrospinal fluid), modulation and spatial smoothing of images with 8-mm full-width at half-maximum isotropic Gaussian kernel.

We compared data between the CO and control groups. Age and sex were considered covariates of no interest to exclude their possible effects on regional grey or white matter volume.22 The significance threshold was set at *p* < 0.05 and corrected for multiple comparisons across the whole brain (family-wise error).

### Tract-based spatial statistics

The FA data from all participants were aligned into a common space using a nonlinear image registration toolkit.23 We calculated FA from realigned diffusion tensor and then this projected onto a mean FA skeleton; the resulting data were used for voxel-wise comparisons across participants. We compared the FA data using permutation-based nonparametric inference on cluster size24 and Randomize 2.0 software ([www.fmrib.ox.ac.uk/fsl/randomise/index.html](http://www.fmrib.ox.ac.uk/fsl/randomise/index.html)). We used a restrictive statistical threshold (threshold-free cluster enhancement threshold, *p* < 0.05, corrected for multiple comparisons). We identified abnormal white matter tracts on FA maps based on the atlas prepared at Johns Hopkins University.25

After using the FA images to achieve nonlinear registration and stages of skeleton formation, we estimated the projection vectors from each individual participant onto the mean FA skeleton. The nonlinear warps and skeleton projections were then separately applied to the mean diffusivity, axial and radial eigenvalue images. We analyzed the resulting statistical maps at a threshold of *p* < 0.05 (corrected at cluster level for multiple comparisons).

### Statistical analysis

We used the Kruskal–Wallis *H* test to compare neuropsychiatric performance between those in the CO and control groups because these data were not normally distributed. We used a paired *t* test to test for differences between the performance of patients at 3 and 10 months.

We performed further analyses on the axial and radial eigenvalues extracted from the aligned database for the regions showing differences in FA values between the CO and control groups. We evaluated the relation between diffusion parameters (FA, mean diffusivity, axial and radial eigenvalues) and cognitive domains in the CO group using nonparametric correlation analysis. The *r* values within the regions of interest against the cognitive scores were obtained by use of SPSS (version 11.0 for Windows). We considered *p* < 0.01 statistically significant.

## Results

### Patients

The CO group comprised 4 male and 5 female patients (Table 1). Patients were diagnosed with CO intoxication based on a history of a charcoal burning suicide attempt and an elevated carboxyhemoglobin level (> 10%) at the emergency department (mean 20.4%, range 15%–25%). All patients had an initial loss of consciousness and received normobaric oxygen therapy. None had a neurologic or neuropsychiatric history. Brain MRI and neuropsychologic examinations were performed after 3 months (mean 86, standard deviation [SD] 8, d) and 10 months (mean 312, SD 17, d).

View this table:
[Table 1](http://jpn.ca/content/35/2/115/T1)

Table 1 
Characteristics of the 9 carbon monoxide–intoxicated patients

### Cognitive tests and clinical features

The cognitive data for CO patients and controls is summarized in Table 2. The CO group scores at both the 3- or 10-month follow-up visit were lower than those of controls in multiple domains, including visual and verbal memory, executive function, calculation and visuospatial function. Paired *t* tests did not reveal significant differences between the 3- and 10-month follow-up data in the CO group.

View this table:
[Table 2](http://jpn.ca/content/35/2/115/T2)

Table 2 
Neuropsychologic comparisons of carbon monoxide–intoxicated patients 3 and 10 months after exposure and control participants

The CO group at 3 months had higher depression scores on NPI than controls (4.9 v. controls 1.45, *p* = 0.01, *t*25 = −2.79) and NPI total score (*p* = 0.004, *t*25 = −3.13; Table 1). The scores in the CO group were not significantly different from the control scores at 10 months. In the CO group, no NPI domain score was significantly different between 3 and 10 months.

### Diffusion parameters at 3 months

The red voxels in Figure 1(A–D) show the areas where FA was reduced in the CO group. These regions include the anterior and posterior corpus callosum, orbitofrontal cortex, internal capsule, anterior region of the external capsule, thalamus and corona radiata. These regions overlapped with the regions with increased mean diffusivity in the CO group (Fig. 1E–H). Panels I and J in Figure 1 represent regions showing increased axial eigenvalue in the CO group, which are the anterior part of the external capsule and frontal corona radiata. Regions with increased radial eigenvalue overlapped with FA and mean diffusivity regions. No regions showed decreased FA, increased mean diffusivity, decreased axial eigenvalue or increased radial eigenvalue in the control group.

![Fig. 1](http://jpn.ca/https://www.jpn.ca/content/jpn/35/2/115/F1.medium.gif)

[Fig. 1](http://jpn.ca/content/35/2/115/F1)

Fig. 1 
Clusters with significant differences (red) between the carbon monoxide group 3 months after exposure and controls. The clusters were overlaid on the mean fractional anisotropy skeleton map (green) (all *p* < 0.05, multiple correction at the cluster level). The background anatomic reference was the Montreal Neurological Institute 152 Image.

### Diffusion parameters at 10 months

Figure 2(A–D) shows the regions where FA was reduced in the CO group compared with controls, including the anterior and posterior corpus callosum, orbitofrontal cortex, internal capsule, anterior region of external capsule, thalamus and corona radiata. Regions showing increased mean diffusivity in the CO group (Fig. 2E–H) were more extensive than the FA map, which showed additional regions, such as cerebral peduncles, brain stem, cerebellar vermis and subcortical U fibres. Regions with increased axial eigenvalue in the CO group overlapped with regions of increased mean diffusivity (Fig. 2I–M), where there was more extensive cerebellar hemisphere white matter involvement (Fig. 2I). Regions with increased radial eigenvalue in the CO group (Fig. 2N–Q) overlapped with the mean diffusivity map, except with less involvement of the brain stem region. No regions showed decreased FA, increased mean diffusivity, decreased axial eigenvalue or increased radial eigenvalue in the control group.

![Fig. 2](http://jpn.ca/https://www.jpn.ca/content/jpn/35/2/115/F2.medium.gif)

[Fig. 2](http://jpn.ca/content/35/2/115/F2)

Fig. 2 
Clusters with significant differences (red) between the carbon monoxide group 10 months after exposure and controls. The clusters were overlaid on the mean fractional anisotropy skeleton map (green) (all *p* < 0.05, multiple correction at the cluster level). The background anatomic reference was the Montreal Neurological Institute 152 Image.

### Diffusion parameters at 3 and 10 months

Regions showing relatively decreased FA at 3 months (Fig. 3A–C) included the inferior fronto-occipital and inferior longitudinal fasciculus, corticospinal tract and superior longitudinal fasciculus. Areas with increased mean diffusivity at 10 months (Fig. 3D–G) included the anterior and posterior corpus callosum, right thalamus, superior longitudinal fasciculus and inferior frontal occipital fasciculus. Panels H–L in Figure 3 show regions with decreased axial eigenvalue at 3 months, including the brain stem, anterior and posterior corpus callosum, orbitofrontal cortex, internal capsule, anterior region of external capsule, thalamus and corona radiata. No region showed differences in radial eigenvalue between 3 and 10 months, and no region showed decreased FA, mean diffusivity or axial eigenvalue at 10 months compared with at 3 months.

![Fig. 3](http://jpn.ca/https://www.jpn.ca/content/jpn/35/2/115/F3.medium.gif)

[Fig. 3](http://jpn.ca/content/35/2/115/F3)

Fig. 3 
Clusters with significant differences (red) between the 3- and 10-month follow-up images in the carbon monoxide groups overlaid on the mean fractional anisotropy skeleton map (green) (all *p* < 0.05, multiple correction at the cluster level). The background anatomic reference was the Montreal Neurological Institute 152 Image.

Values of axial and radial eigenvalues were further extracted from the regions where FA was reduced at 3 months for further analysis (Fig. 4). The results showed that reduced FA resulted from changes of axial eigenvalue (Fig. 4A) rather than radial eigenvalue (Fig. 4B).

![Fig. 4](http://jpn.ca/https://www.jpn.ca/content/jpn/35/2/115/F4.medium.gif)

[Fig. 4](http://jpn.ca/content/35/2/115/F4)

Fig. 4 
Bar graph showing eigenvalues extracted from the significant regions between the 3- and 10-month follow-up images in the carbon monoxide (CO) group. (**A**) Axial eigenvalues in CO group show significant differences in the 6 regions (*p* < 0.05) between 3 and 10 months. (**B**) Values in radial eigenvalues were not different between 3 and 10 months in the CO group. Axial and radial eigenvalues are expressed in 103 mm2/s. Data are means and standard deviations. T-P = temporoparietal.

### Voxel-based morphometry

Comparison between the CO group (at either 3 or 10 months) and the control group showed no voxels with significant differences in grey matter volume. A direct comparison between 3 and 10 months in the CO group also did not reveal any areas with significant differences.

Comparison between the CO group at 3 months and the control group did not reveal significant differences in white matter regions. However, there was significant atrophy in the periventricular white matter at 10 months when the CO group was compared with the control group (Fig. 5). These regions included the anterior and posterior corpus callosum and the white matter tract in the temporal-parietal subcortical regions. There were no significant differences in white matter volume between 3 and 10 months in the CO group.

![Fig. 5](http://jpn.ca/https://www.jpn.ca/content/jpn/35/2/115/F5.medium.gif)

[Fig. 5](http://jpn.ca/content/35/2/115/F5)

Fig. 5 
Atrophy of white matter in carbon monoxide–intoxicated patients 10 months after exposure compared with controls by use of voxel-based morphometry. The anatomic reference was the white matter template specific to this study.

### Correlation between neuropsychologic testing and diffusion parameters

#### Relation between FA and cognitive domains

At 3 months, correlation between FA and cognitive domains showed that the MMSE score was correlated with left temporal parietal white matter (*r* = 0.75), bilateral high frontal regions (*r* left = 0.728, right = 0.767). The CDR score was inversely correlated with left temporal parietal white matter (*r* = −0.894), left lower corona radiata (*r* = −0.827) and right higher corona radiata (*r* = −0.7). Digit backward scores correlated with left temporal parietal white matter (*r* = 0.789) and right high corona radiata (*r* = 0.77), whereas verbal generation score correlated with left inferior frontal (*r* = 0.804), high frontal (*r* = 0.751), temporal parietal white matter (*r* = 0.718) and right lower corona radiata (*r* = 0.76). Delayed memory scores correlated with left temporal parietal white matter (*r* = 0.806) and left lower corona (*r* = 0.786).

At 10 months, the digit backward scores were correlated with FA in the left inferior frontal (*r* = 0.894). Verbal generation score were correlated with left temporal parietal white matter (*r* = 0.975).

#### Relation between mean diffusivity and cognitive domains

At 3 months, both MMSE (*r* = −0.85) and CDR (*r* = 0.783) scores correlated with the mean diffusivity value of the right inferior frontal region. Scores of judgment were inversely correlated with the right high frontal mean diffusivity value (*r* = −0.675).

At 10 months, verbal fluency scores were inversely correlated with the right (*r* = −0.975) and left (*r* = −0.975) external capsule.

#### Relation between axial eigenvalues and cognitive domains

No domains showed correlation at 3 months with axial eigenvalues. At 10 months, axial diffusivity was inversely correlated with digit backward scores (*r* = −0.894), and the delay recall memory correlated with the right lower corona radiata (*r* = −0.89).

#### Relation between radial eigenvalues and cognitive domains

At 3 months, MMSE (*r* = −0.833) and CDR (*r* = 0.894) scores were highly correlated with radial eigenvalues. Digit backward score were inversely correlated with the left high corona radiata (*r* = −0.912). Verbal fluency scores were inversely correlated with the left lower corona radiata (*r* = −0.73), right lower corona radiata (*r* = −0.876) and left temporal parietal white matter (*r* = −0.838), whereas scores of calculation inversely correlated with the left (*r* = −0.868) and right corona radiata (*r* = −0.806). At 10 months, delay memory scores were inversely correlated with the left lower (*r* = −0.89) and high corona radiata (*r* = −0.89).

## Discussion

Our study suggests that there were increased changes over time in the CO group, which is consistent with previous findings26–28 that neural damage continues variably between patients and structures. Furthermore, our findings may also provide a possible mechanism. Because diffusion indices such as FA and mean diffusivity are derivatives from the 3 tensor eigenvalues, our analysis corroborates the changes contributed by each eigenvalue and shows that extensive changes in the FA map at both 3 and 10 months in the CO group are contributed to initially by increases in radial eigenvalues. This longitudinal study shows that axial eigenvalues evolve during the follow-up period with accompanying atrophy of the white matter tract, which supports the results of increased mean diffusivity in the CO group.

Using an apparent diffusion coefficient map, Kim and colleagues29 showed that reduced diffusivity developed in the white matter 25 to 95 days after CO intoxication; they interpreted this as slowly progressive cytotoxic edema.30 The time window studied by Kim and colleagues29 was earlier than in our study. Our study and one by Terajima and colleagues31 found a gradual increase of mean diffusivity after 3 months following CO exposure. In conjunction, values of mean diffusivity may initially be lower than controls and then increase, reaching a level higher than controls. A longitudinal increase in mean diffusivity suggests less restriction of water diffusion in all directions. One explanation for these changes may be tissue atrophy resulting from unrecovered white matter damage.26,31 Apparently, mean diffusivity values detected the changes earlier because our VBM result also detected the volume reduction after 10 months of follow-up.

This study also shows increases in radial eigenvalues at 3 months. One explanation for this phenomenon is that more free water diffuses across the myelin barrier, which may correspond to myelin damage.32 Because there were no interval changes in radial eigenvalues, our results suggest that disruption of myelin is persistent and recovery from damage does not occur after 10 months. From axial diffusivity analysis, only the anterior part of the external capsules and frontal corona radiata are affected at 3 months, with further increases in axial diffusivity in the deep white matter at 10 months. This indicates enlargement of affected regions by axonal injury from CO toxicity.33,34 Based on the changes of axial and radial eigenvalues, increasing FA values from 3 to 10 months are attributed to the axial eigenvalues changes, which should not be misinterpreted as recovery of fibre coherence.

From our comparison with age-matched controls, we can see that cognitive presentation, as quantified by neuropsychologic data, supports the cognitive sequel of CO. Cognitive presentations in CO intoxication can be partly explained by reduced connectivity between different cortical regions.35,36 Compared with the control group, the CO group had lower scores in digit backward and verbal fluency, which may be related to interruption of the frontotemporal circuits.37–39 Impairment in free recall rather than retrieval suggests that memory deficits may be due to subcortical deficits,40 whereas interruption of the temporal visual pathway results in visuospatial deficits.

From direct comparison of neuropsychologic data from the CO group at 3 and 10 months after exposure, it was evident that the recovery of cognition was not significant. In conjunction with analysis of diffusion parameters and atrophy in white matter, we speculate that the test scores at 10 months after exposure should be worse than at 3 month if white matter damage is a major predictor of outcome. Although the major white matter atrophy in this study was in the periventricular white matter, one possibility is that periventricular white matter is not a major determinant of cognitive deficits in CO encephalopathy. This is based on our cognitive test with FA correlation results, which showed that the semicentrum ovale rather than the deep white matter had better correlation with cognitive decline. This was also observed by Parkinson and colleagues,41 suggesting that the semicentrum ovale, not the periventricular areas, is significantly associated with cognitive decline. Another explanation is that even though white matter changes are commonly observed in CO encephalopathy, interpretation of neuropsychologic deficits should also consider the effects of cortical involvement. From the grey matter VBM results, longitudinal atrophy in grey matter was not different from age-matched controls. The relative sparing of axonopathy on subcortical U fibres without grey matter atrophy may account for the undetected cognitive deterioration at 10 months in this study.

### Limitations

There are several limitations to this study. First, we enrolled a small number of patients. A larger study group may be necessary to confirm our observations. Second, we cannot provide histologic proof in this study because this was a longitudinal study involving humans. Therefore, the interpretation of the pathological changes through neuroimaging findings (as reflected by changes in axial and radial eigenvalues) cannot be proved. Nonetheless, the results are consistent with those of previous pathology studies in which myelin damage was constant with or without axonal destruction.30,42 It is worth noting that a direct correlation between axial and radial diffusivity parameters and the microscopic pathology of white matter injury has not been conclusive. As such, interpretation of changes in axial or radial eigenvalues as demyelination or axonopathy43,44 should be treated with great care especially in regions where neuron fibres cross.45

## Conclusion

This in vivo human study showed the evolution of white matter diffusion changes and atrophy regions after CO intoxication. Diffusion measures were better correlated with cognitive decline on the fibre tract of the semicentrum ovale and not the periventricular fibres. Although changes in grey matter volume were not evident, neuropsychiatric deficits were persistent in the follow-up study, suggesting the importance of white matter integrity.

## Acknowledgements

The authors thank Dr. C.F. Jiang, I-Shou University, for support with DTI acquisition and analyses, and Dr. C.W. Chiang, Kaohsiung Naval General Hospital, for referring patients with CO intoxication. This work was supported in part by grant CM-RPG860171 from the Chang Gung Memorial Hospital.

## Footnotes

*   **Competing interests:** Dr. Wang has received speakers fees from Johnson & Johnson, Taiwan. None declared for all other authors.

*   **Contributors:** Drs. C.-C. Chang, W.-N. Chang and S.-S. Chen designed the study. Drs. C.-C. Chang, Lui, C.-F. Chen, Lee, S.-S. Chen, Lin, Huang and C. Chen acquired the data, which Drs. C.-C. Chang, Wang and S.-S. Chen analyzed. Drs. C.-C. Chang, Lui and Wang wrote the article. Drs. W.-N. Chang, Lui, Wang, C.-F. Chen, Lee, S.-S. Chen, Lin, Huang and C. Chen reviewed it. All authors approved the article’s final publication.

*   Received May 12, 2009.
*   Revision received September 14, 2009.
*   Revision received November 2, 2009.
*   Accepted November 3, 2009.

## References

1.  Kuo CJ, Conwell Y, Yu Q, et al.Suicide by charcoal burning in Taiwan: implications for means substitution by a case-linkage study.Soc Psychiatry Psychiatr Epidemiol 2008;43:286–90.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1007/s00127-007-0300-2&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=18075708&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

2.  Liu KY, Beautrais A, Caine E, et al.Charcoal burning suicides in Hong Kong and urban Taiwan: an illustration of the impact of a novel suicide method on overall regional rates.J Epidemiol Community Health 2007;61:248–53.
    
    [Abstract/FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiamVjaCI7czo1OiJyZXNpZCI7czo4OiI2MS8zLzI0OCI7czo0OiJhdG9tIjtzOjE4OiIvanBuLzM1LzIvMTE1LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

3.  Young LJ, Caughey WS.Pathobiochemistry of CO poisoning.FEBS Lett 1990;272:1–6.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/0014-5793(90)80436-M&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=2172009&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1990EF04400001&link_type=ISI) 

4.  Weaver LK.Clinical practice. Carbon monoxide poisoning.N Engl J Med 2009;360:1217–25.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1056/NEJMcp0808891&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=19297574&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000264283400008&link_type=ISI) 

5.  Hopkins RO, Fearing MA, Weaver LK, et al.Basal ganglia lesions following carbon monoxide poisoning.Brain Inj 2006;20:273–81.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1080/02699050500488181&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16537269&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000235972400007&link_type=ISI) 

6.  Hopkins RO, Woon FL.Neuroimaging, cognitive, and neurobehavioral outcomes following carbon monoxide poisoning.Behav Cogn Neurosci Rev 2006;5:141–55.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1177/1534582306289730&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=16891556&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

7.  Weaver LK, Valentine KJ, Hopkins RO.Carbon monoxide poisoning: risk factors for cognitive sequelae and the role of hyperbaric oxygen.Am J Respir Crit Care Med 2007;176:491–7.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1164/rccm.200701-026OC&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17496229&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000249173700013&link_type=ISI) 

8.  Ringman JM, O’Neill J, Geschwind D, et al.Diffusion tensor imaging in preclinical and presymptomatic carriers of familial Alzheimer’s disease mutations.Brain 2007;130:1767–76.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1093/brain/awm102&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17522104&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000248151300007&link_type=ISI) 

9.  Le Bihan D.Looking into the functional architecture of the brain with diffusion MRI.Nat Rev Neurosci 2003;4:469–80.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1038/nrn1119&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12778119&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000183221000016&link_type=ISI) 

10. Pierpaoli C, Basser PJ.Toward a quantitative assessment of diffusion anisotropy.Magn Reson Med 1996;36:893–906.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1002/mrm.1910360612&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=8946355&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1996VU89000011&link_type=ISI) 

11. Sykova E.Diffusion properties of the brain in health and disease.Neurochem Int 2004;45:453–66.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.neuint.2003.11.009&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=15186911&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000222295500002&link_type=ISI) 

12. Rorschach HE, Lin C, Hazlewood CF.Diffusion of water in biological tissues.Scanning Microsc Suppl 1991;5:S1–10.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=1822018&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1991JC39500029&link_type=ISI) 

13. Kraus MF, Susmaras T, Caughlin BP, et al.White matter integrity and cognition in chronic traumatic brain injury: a diffusion tensor imaging study.Brain 2007;130:2508–19.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1093/brain/awm216&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17872928&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000250039500005&link_type=ISI) 

14. Folstein MF, Folstein SE, McHugh PR.“Mini-Mental State.” A practical method for grading the cognitive state of patients for the clinician.J Psychiatr Res 1975;12:189–98.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/0022-3956(75)90026-6&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=1202204&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1975AY27900005&link_type=ISI) 

15. Schafer KA, Tractenberg RE, Sano M, et al.Reliability of monitoring the clinical dementia rating in multicenter clinical trials.Alzheimer Dis Assoc Disord 2004;18:219–22.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=15592134&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000225930000012&link_type=ISI) 

16. Rosen WG, Mohs RC, Davis KL.A new rating scale for Alzheimer’s disease.Am J Psychiatry 1984;141:1356–64.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1176/ajp.141.11.1356&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=6496779&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1984TQ68200003&link_type=ISI) 

17. Boone KB.The Boston Qualitative Scoring System for the Rey-Osterrieth Complex Figure.J Clin Exp Neuropsychol 2000;22:430–4.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1076/1380-3395(200006)22:3;1-V;FT430&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10979748&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

18. Kaplan EF, Goodglass H, Weintraub S (1983) The Boston Naming Test (Lea and Febiger, Philadelphia (PA)).
    
    

19. Warrington EK, James M (1991) Visual Object and Space Perception Battery (Thames Valley Test Co, Bury St. Edmunds (UK)).
    
    

20. Amieva H, Lafont S, Rouch-Leroyer I, et al.Evidencing inhibitory deficits in Alzheimer’s disease through interference effects and shifting disabilities in the Stroop test.Arch Clin Neuropsychol 2004;19:791–803.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.acn.2003.09.006&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=15288332&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000223414500007&link_type=ISI) 

21. Cummings JL, Mega M, Gray K, et al.The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia.Neurology 1994;44:2308–14.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1212/WNL.44.12.2308&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=7991117&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

22. Good CD, Johnsrude IS, Ashburner J, et al.A voxel-based morphometric study of ageing in 465 normal adult human brains.Neuroimage 2001;14:21–36.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1006/nimg.2001.0786&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=11525331&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000169498000003&link_type=ISI) 

23. Rueckert D, Sonoda LI, Hayes C, et al.Nonrigid registration using free-form deformations: application to breast MR images.IEEE Trans Med Imaging 1999;18:712–21.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1109/42.796284&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=10534053&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000083326700006&link_type=ISI) 

24. Nichols TE, Holmes AP.Nonparametric permutation tests for functional neuroimaging: a primer with examples.Hum Brain Mapp 2002;15:1–25.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1002/hbm.1058&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=11747097&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000172887500001&link_type=ISI) 

25. Wakana S, Jiang H, Nagae-Poetscher LM, et al.Fiber tract-based atlas of human white matter anatomy.Radiology 2004;230:77–87.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1148/radiol.2301021640&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=14645885&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000187613100012&link_type=ISI) 

26. Porter SS, Hopkins RO, Weaver LK, et al.Corpus callosum atrophy and neuropsychological outcome following carbon monoxide poisoning.Arch Clin Neuropsychol 2002;17:195–204.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/S0887-6177(00)00110-4&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=14589748&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

27. Chambers CA, Hopkins RO, Weaver LK, et al.Cognitive and affective outcomes of more severe compared to less severe carbon monoxide poisoning.Brain Inj 2008;22:387–95.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1080/02699050802008075&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=18415719&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000255986500004&link_type=ISI) 

28. Pulsipher DT, Hopkins RO, Weaver LK.Basal ganglia volumes following CO poisoning: a prospective longitudinal study.Undersea Hyperb Med 2006;33:245–56.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17004411&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000240336500005&link_type=ISI) 

29. Kim JH, Chang KH, Song IC, et al.Delayed encephalopathy of acute carbon monoxide intoxication: diffusivity of cerebral white matter lesions.AJNR Am J Neuroradiol 2003;24:1592–7.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=13679276&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000185400100020&link_type=ISI) 

30. Prockop LD, Chichkova RI.Carbon monoxide intoxication: an updated review.J Neurol Sci 2007;262:122–30.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.jns.2007.06.037&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17720201&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000250244900020&link_type=ISI) 

31. Terajima K, Igarashi H, Hirose M, et al.Serial assessments of delayed encephalopathy after carbon monoxide poisoning using magnetic resonance spectroscopy and diffusion tensor imaging on 3.0T system.Eur Neurol 2008;59:55–61.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1159/000109262&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17917459&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000251436400009&link_type=ISI) 

32. Lin Y, Wang J, Wu C, et al.Diffusion tensor imaging of the auditory pathway in sensorineural hearing loss: changes in radial diffusivity and diffusion anisotropy.J Magn Reson Imaging 2008;28:598–603.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1002/jmri.21464&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=18777540&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

33. Roosendaal SD, Geurts JJ, Vrenken H, et al.Regional DTI differences in multiple sclerosis patients.Neuroimage 2009;44:1397–403.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.neuroimage.2008.10.026&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=19027076&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000263861800018&link_type=ISI) 

34. Arfanakis K, Haughton VM, Carew JD, et al.Diffusion tensor MR imaging in diffuse axonal injury.AJNR Am J Neuroradiol 2002;23:794–802.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12006280&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000175599300012&link_type=ISI) 

35. Bucur B, Madden DJ, Spaniol J, et al.Age-related slowing of memory retrieval: contributions of perceptual speed and cerebral white matter integrity.Neurobiol Aging 2008;29:1070–9.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.neurobiolaging.2007.02.008&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17383774&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000256457000012&link_type=ISI) 

36. Markowitsch HJ, Calabrese P, Liess J, et al.Retrograde amnesia after traumatic injury of the fronto-temporal cortex.J Neurol Neurosurg Psychiatry 1993;56:988–92.
    
    [Abstract/FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiam5ucCI7czo1OiJyZXNpZCI7czo4OiI1Ni85Lzk4OCI7czo0OiJhdG9tIjtzOjE4OiIvanBuLzM1LzIvMTE1LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

37. Kiernan JA, Hudson AJ.Frontal lobe atrophy in motor neuron diseases.Brain 1994;117:747–57.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1093/brain/117.4.747&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=7922462&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=A1994PQ94800010&link_type=ISI) 

38. Moretti R, Torre P, Antonello RM, et al.Neuropsychological evaluation of late-onset post-radiotherapy encephalopathy: a comparison with vascular dementia.J Neurol Sci 2005229–230195–200.
    
    

39. Suwanwela N, Srikiatkhachorn A, Tangwongchai S, et al.Mutation of the Notch 3 gene in a Thai cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy family.J Med Assoc Thai 2003;86:178–82.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12678157&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

40. Albert ML, Feldman RG, Willis AL.The “subcortical dementia” of progressive supranuclear palsy.J Neurol Neurosurg Psychiatry 1974;37:121–30.
    
    [Abstract/FREE Full Text](http://jpn.ca/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiam5ucCI7czo1OiJyZXNpZCI7czo4OiIzNy8yLzEyMSI7czo0OiJhdG9tIjtzOjE4OiIvanBuLzM1LzIvMTE1LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

41. Parkinson RB, Hopkins RO, Cleavinger HB, et al.White matter hyperintensities and neuropsychological outcome following carbon monoxide poisoning.Neurology 2002;58:1525–32.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1212/WNL.58.10.1525&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12034791&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

42. Funata N, Okeda R, Takano T, et al.Electron microscopic observations of experimental carbon monoxide encephalopathy in the acute phase.Acta Pathol Jpn 1982;32:219–29.
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=7090814&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 

43. Schmierer K, Wheeler-Kingshott CA, Boulby PA, et al.Diffusion tensor imaging of post-mortem multiple sclerosis brain.Neuroimage 2007;35:467–77.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1016/j.neuroimage.2006.12.010&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=17258908&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000245293100003&link_type=ISI) 

44. Song SK, Sun SW, Ramsbottom MJ, et al.Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water.Neuroimage 2002;17:1429–36.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1006/nimg.2002.1267&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=12414282&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000179012800031&link_type=ISI) 

45. Wheeler-Kingshott CA, Cercignani M.About “axial” and “radial” diffusivities.Magn Reson Med 2009;61:1255–60.
    
    [CrossRef](http://jpn.ca/lookup/external-ref?access_num=10.1002/mrm.21965&link_type=DOI) 
    
    [PubMed](http://jpn.ca/lookup/external-ref?access_num=19253405&link_type=MED&atom=%2Fjpn%2F35%2F2%2F115.atom) 
    
    [Web of Science](http://jpn.ca/lookup/external-ref?access_num=000265566000032&link_type=ISI)