| Objective To evaluate the feasibility of QSM in measurement of brain iron deposition in normal subjects and patients with haemodialysis. First, to evaluate the distribution and age-related changes of brain iron content in healthy controls using QSM at3.0T magnetic resonance imaging (MRI) system. Second, to compare the magnetic susceptibility of different brain structures in patients with haemodialysis to that in age-and sex-matched healthy controls using QSM and explore the correlations between susceptibility of different brain structures in patients with haemodialysis and neuropsychiatric test scores, and epidemiological and clinical blood biochemical parameters.Methods First, from November2012to July2013,63healthy right-handed volunteers (male24, female39, age20-63years, mean37±13years) were collected from Tianjin local residents. All healthy subjects underwent the conventional MRI and unfiltered SWI scan to get the unfiltered phase and magnitude images using3.0T, SM was reconstructed from the unfiltered phase and magnitude images by Susceptibility Mapping and Phase Artifacts Removal Toolbox (SMART) software. The regions of interest of bilateral frontal white matter and deep gray matter nuclei were outlined manually and the susceptibility was measured by Signal Processing in Nuclear Magnetic Resonance (SPIN) software. Spearman correlation analysis was used to calculate the correlation between the susceptibility and iron concentration cited from two post-mortem brain studie. Wilcoxon test was applied to compare susceptibility of bilateral frontal white matter and deep gray matter nuclei. Spearman correlation analysis was used to calculate the correlation between the susceptibility of frontal white matter and deep gray matter nuclei and age. Then, Fifty three patients with haemodialysis (31male and22female, mean age47.4±13.5years) with median duration of haemodialysis of26.7months (1to114months) and45age-and sex-matched healthy controls (18male and27female, mean age42.3±11.7years) were recruited in this retrospective study from our hospital and Tianjin local residents, respectively. All subjects were right-handed and had a conventional MRI and unfiltered SWI scan on3.0T MRI system. QSM was reconstructed from unfiltered phase data. Susceptibility was measured in the bilateral frontal white matter and deep cerebral gray matter nuclei on QSM and used to compare between patients with haemodialysis and healthy controls by two independent samples t-test. The difference between neuropsychiatric test score of patients with haemodialysis and healthy subjects was calculated by two independent samples t-test. The correlation between susceptibility of different brain structures of patients with haemodialysis and neuropsychiatric test score was calculated using Spearman correlation analysis. The correlation between susceptibility of different brain structures of patients with haemodialysis and epidemiological and clinical blood biochemical parameters were calculated by Step wise multiple regression analysis.Results There was a strong positive correlation between the susceptibility and iron concentration gathered from two post-mortem brain studies (r=0.905,P=0.002; r=0.900, P=0.037):(1) In the healthy subjects, the susceptibility of globus pallidus was the highest, the second was substantia nigra, the least susceptibility was seen in frontal white matter;(2) In the healthy subjects, there was significant difference between susceptibility of bilateral caudate head, substantia nigra, red nucleus and dentate nucleus (P<0.05);(3) The gender-related difference in susceptibility of bilateral globus pallidus between male and female was statistically significant, the left P=0.024, the right P=0.016;(4) There was a positive correlation between age and susceptibility of bilateral caudate head, putamen, red nucleus and dentate nucleus (P<0.05); the mean susceptibility of these nuclei also had a positive correlation with age (P<0.05);(5) In patients with haemodialysis, the susceptibility of the bilateral caudate head, putamen, substantia nigra, red nucleus and dentate nucleus was statistically higher than that in healthy controls (P<0.01);(6) The susceptibility of the left caudate head (r=-0.415) and right caudate head (r=-0.311) had negative correlation with neuropsychiatric test scores (all P<0.05);(7) In the patients with haeodialysis, we found that there were correlations between the susceptibility of some brain structures and epidemiological and clinical blood biochemical parameters. The duration of dialysis was an independent factor to the increaed susceptibility of bilateral PUT and RTHA; cholesterol was identified as an independent factor to the increased susceptibility of LCA; unsaturated iron-bonding capacity (UIBC) was identified as an independent factor to the increased susceptibility of LGP; post-dialysis DBP was an independent factor to the increased susceptibility of RGP;β2-MG was an independent factor to the increased susceptibility of LTHA; post-dialysis PP was an independent factor to the increased susceptibility of RSN; HB was an independent factor to the increased susceptibility of RDN; and C-reactive protein was correlated to the increased susceptibility of RFWM (all P<0.05);(8) There was no correlation between susceptibility of any brain structures of patients with haemodialysis and serum ferritin (P>0.05).Conclusion QSM can accurately evaluate the brain iron deposition, which appears high signal on QSM and is correlated with iron concentration from two post-mortem brain studies and brain iron deposition of bilateral caudate head, putamen, red nucleus and dentate nucleus increase with aging. Our findings indicated that increased brain iron deposition do occur in patients with haemodialysis, which is positively correlated with the duration of dialysis. The increased susceptibility of the caudate nuclei appeared to reflect neurocognitive impairment, especially executive function and the susceptibility may be a clinical biomarker to reflect the iron changes in the brain. |
PDF Full Text Request