Diffusion tensor imaging in type 1 diabetes: decreased white matter integrity relates to cognitive functions
- 1.1k Downloads
KeywordsBrain Cognition Imaging Microangiopathy Type 1 diabetes White matter tracts
Diffusion tensor imaging
Tract-based spatial statistics
To the Editor: Type 1 diabetes, particularly in the presence of microangiopathy, is associated with cognitive dysfunction, mainly observed in domains involving processing speed, suggesting white matter involvement . White matter hyperintensities, a commonly used marker for white matter damage on MRI, however, do not occur more prevalently in type 1 diabetes compared with controls . Therefore, we assessed white matter tract integrity using MRI-diffusion tensor imaging (DTI) and cognitive functions in type 1 diabetic patients with and without microangiopathy and in controls. We hypothesised that type 1 diabetic patients with microangiopathy would show the most pronounced reductions in white matter tract integrity compared with the other groups, and that these differences would be associated with cognitive differences.
Forty-eight patients with microangiopathy (mean age: 44.6 ± 7.3 years; mean HbA1c: 8.1 ± 1.3% [65.1 ± 14.4 mmol/mol]; mean disease onset age: 10.3 ± 7.1 years; mean disease duration: 34.3 ± 7.9 years; microalbuminuria: 29%; self-reported neuropathy: 52%), 52 patients without microangiopathy (age: 38.1 ± 9.1 years; HbA1c: 7.8 ± 0.9% [61.6 ± 9.9 mmol/mol]; disease onset age: 16.4 ± 9.6 years; disease duration: 21.7 ± 9.3 years) and 49 controls (age: 36.7 ± 11.2 years; HbA1c: 5.3 ± 0.2% [34.2 ± 2.6 mmol/mol]), matched for sex, IQ and BMI, were included. Inclusion and exclusion criteria, together with definitions of microangiopathy and severe hypoglycaemia, have been previously published . Patients with microangiopathy were selected if they had proliferative retinopathy, but could also have other complications . Those without microangiopathy had no clinically measurable complications. All participants filled out the Centre of Epidemiological Studies Scale of Depression , and routine blood and urine sampling was performed. Blood glucose during testing was kept between 4 and 15 mmol/l.
The neuropsychological assessment covered the following domains: general cognitive ability, memory, information processing speed, executive functions, attention, and motor and psychomotor speed . MRI scanning was performed at 1.5 T (Siemens Sonata, Erlangen, Germany). DTI acquisition consisted of 10 volumes without directional weighting and 60 volumes with 60 non-collinear gradient directions (b-value 700 s/mm2), repetition time 8500 ms; echo time 86 ms; 59 contiguous axial slices, isotropic 2 mm resolution. DTI post-processing with FSL4.1 provided eigenvectors λ1, λ2 and λ3, and the derived parameters fractional anisotropy ([FA] general white matter integrity) and axial (diffusion parallel to the axon), radial (diffusion perpendicular to the axon) and mean (overall diffusion) diffusivity . Tract-based spatial statistics (TBSS) were applied for FA , and for axial, radial and mean diffusivity. Voxel-based statistics with ‘randomise’ were corrected for multiple comparisons using the family-wise error (FWE) . In case of an effect in all patients vs. controls, post hoc tests were performed with individual patient groups. Tractography was used to determine diffusion parameters in the bilateral corticospinal and inferior fronto-occipital tracts, as these showed the largest differences between patients and controls. Correlations between cognition and DTI parameters of these tracts were determined using linear regression. All analyses were corrected for age, sex, systolic blood pressure and depressive symptoms. A p value <0.05 (FWE-corrected) was considered to be statistically significant.
The current FA results are comparable with one earlier small-sized DTI-study in type 1 diabetic patients , but, in addition, we showed that FA decrease is more extensive in patients with microangiopathy. We have extended these observations by the finding that spatially widespread reductions in axial diffusivity already occur in patients without microangiopathy and are most marked in those with microangiopathy, suggesting that axonal damage or loss of coherence in fibre bundles is an early process in type 1 diabetes . The observed increase in radial diffusivity, occurring as microangiopathy develops, is thought to represent myelin damage . As studies on the biological correlates of axial and radial diffusivity are performed in animals and not humans, the interpretation is still speculative. The current results do not support involvement of severe hypoglycaemia in these white matter tract changes. Simultaneous assessment of grey and white matter changes, as previously performed , would be an interesting future direction. Better cognitive performance was correlated with better white matter tract integrity in both patient groups. Longitudinal studies need to identify the course and (hyperglycaemia-related) underlying mechanisms of these diffusion changes as diabetes progresses.
This study was supported by Grant 2006.00.006 of the Dutch Diabetes Research Foundation.
EvD, MD, FJS, MK and FB participated in the conception and design of the study. EvD performed the study, MRI and statistical analyses and wrote the article. MMS and PJWP supervised the MRI analyses. RGIJ supervised the statistical analyses. ACM rated all fundus photographs. CMR critically reviewed and interpreted the data. All authors were involved in interpretation of the data, drafted the text and made crucial revisions to the article and gave final approval for the article to be published.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
Open AccessThis is an open access article distributed under the terms of the Creative Commons Attribution Noncommercial License (https://creativecommons.org/licenses/by-nc/2.0), which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.