The Research In Coal Mine Disaster Surviors With Post-traumatic Stress Disorder Based On Multi-modality MRI

Post-traumatic stress disorder (PTSD) is an anxiety disorder that developsafter exposure to a terrifying event or ordeal in which grave physical harmoccurred or was threatened. Its features persist for a long duration. The commonsymptoms of PTSD are re-experience trauma, irritability, and avoidance behavior,accompanied by sleep disorders, lack of concentration, high alertness, excessivescare, etc. In general, the symptom is occurred in several days to six months afterexposure to the trauma. It’s reported that about3.6%Americans have sufferedfrom PTSD, and7.8%of them cannot be cured. The PTSD may be induced bywarfare, violent crime, sexual abuse, traffic accident, unemployment, divorce,natural disaster, etc. The patients of PTSD are the survivors, witnesses, andrescue workers of the trauma. At present, PTSD is hot topic of epidemiology,psychology, social medicine. A recent study reported that different types oftraumatic experiences may result in different levels of PTSD severity and displaydistinct PTSD symptom patterns. Especially, in recent years, the coal minedisaster, earthquakes, tsunamis, terrorist attack occur frequently. Therefore, thekind of PTSD induced by the sudden disaster is more and more concerned. Currently, many researches have investigated the pathological andetiological mechanism, according to genetics, nerve, biochemistry, incretion,psychology and sociology studies. However, the detailed mechanism of PTSDremains unclear. Due to lack of clear understanding the alteration of braininduced by PTSD, there is no effectively therapeutic method for PTSD.With the development of medical imaging, it provides a new method to thestudy of brain diseases. Especially, Magnetic Resonance Imaging (MRI) hasgradually developed from structural imaging to functional imaging, fromsingle-modality to multi-modality imaging. To analyze the alteration of structure,blood perfusion, metabolism and connection information in patients with PTSDfrom a range of directions, and combine multi-modality of MRI to diagnosis andevaluate the brain disease, is the hot and difficult topic at present. It is reportedthat different degree of pathological damage was preliminary found in thepatients with PTSD, based on the medical imaging.For psychiatric diseases, the alterations of metabolism, perfusion and theshape of cortical cortex (cortical thickness) are usually occurred before thesubstantive lesions. So the cortical thickness and cerebral blood flow (CBF)analysis are often used to detect the effect of brain disease on the structure andfunction at early stage of disease. Fractional anisotropy (FA) and apparentdiffusion coefficient (ADC) are thought to reflect the microstructural alteration,physiological and pathological changes at the molecular level. However, to ourknowledge, no studies have investigated the effect of PTSD induced by coal minedisaster on the cortical thickness, CBF, FA and ADC value.In view of above questions, the present study analyzed the difference incortical thickness, CBF, FA and ADC of whole brain in patient with recent onsetPTSD due to a mining disaster. The elapsed time between the traumatic event and MRI scans was ranged from187to190days. After the comparison of corticalthickness and CBF between subjects with and without PTSD, the relationship ofcortical thickness and CBF of identified regions with the PTSD symptom severitywas investigated.In2007, a severe coalmine-flooded disaster occurred at Sanmenxia of HenanProvince. Sixty-nine male miners were trapped in a nearly1400m undergroundcoal pit. All the twenty subjects in this study are the survivors from this suddendisaster. Ten of them met the diagnostic criteria for PTSD, while the other tenpersons didn’t meet. With the constructed PTSD and non-PTSD groups in thisstudy, each group had10male subjects. The severity of their symptoms wasassessed with the Chinese version of the Clinician-Administered PTSD Scale(CAPS)1. Cortical thickness measurement and analysisIn this study, the pipeline for cortical thickness measurement and analysiswas proposed, which improved computation speed and the accuracy of thicknessmeasurement. Considering DARTEL has a higher sensitivity in detecting subtleabnormalities of structures and accurate realignment of subtle structure, in thisstudy, the VBM-DARTEL was used to preprocess the T1sequence of MRI andsegment the normalized data into gray matter, white matter and cerebrospinalfluid (CSF) with signal intensity and prior probability information. Based on thesegmented gray matter, white matter and CSF, the3D cortical thickness wasestimated by the Laplacian approach. According to building series ofequipotential surfaces, the cortical thickness is defined as the arc length of thestreamlines which is perpendicular to the equipotential surfaces. This methodavoided the measurement error from Euclidean distance with unique andreversible corresponding points between two surfaces, and is regarded as one of most accurate and efficient thickness measurement method currently. Due to thetopological shell of the cerebral cortex and various influence factors in dataacquisition, it is hard to classify the tissues perfectly, especially in sulci regions.In this study, the algorithm developed for sulci detection and thickness correctionwas employed to further improve the brain segmentation after the thicknessestimation. After the correction of the segmentation, the cortical thickness wasre-calculated accordingly and more accurate result was obtained for the wholebrain.Based on the image of cortical thickness, the difference of thickness can beanalyzed with statistical parametric mapping (SPM). Group differences indemographic variables were examined with independent t-tests. All subjects inthis study came from the same community and did not differ significantly insocioeconomic status. The two groups (PTSD and non-PTSD) significantlydiffered in age. Thus, group comparisons of cortical thickness were performedusing analysis of two sample t-test in SPM8with age as covariates to account forthe age effect. Comparisons of cortical thickness in survivors experienced miningdisaster indicated that in left parietal lobe, right inferior frontal gyrus, and rightparahippocampal gyrus, cortical thickness of subjects with PTSD was obviouslythinner than those without PTSD.To investigate the relation between thickness and s PTSD symptom severity,the region that consists of all the voxels on the streamline across the significantlythinner region was used for ROI analysis in this study. ROI-based correlationalanalysis was performed in the patients with PTSD. The mean cortical thickness ofright inferior frontal gyrus tended to correlate negatively with the CAPS score inthe identified regions from thickness analysis.2. CBF measurement and analysis According to selection of MRI perfusion sequences, in this study, the pulsedartery spin labeling (PASL) sequence was used to investigate the alteration ofCBF for the patients of PTSD induced by coal mine disaster, which isnon-invasive with high efficiency. In this study, an accurate CBF measurementmethod of PASL sequence was proposed.Due to fast imaging of PASL, the resolution of image is relative low. For thisreason, SPM was used to preprocess the PASL and T1sequences, respectively.With the information of structural images applied to the functional images, theaccuracy of normalization and resolution of functional images were improved.Though the resolution of images had been improved, the images were stillaffected by partial volume effect (PVE). In this study, the acquired time of PASLsequence was used to construct the time sequence, and the MAP-EM algorithmwas used to estimate the mixed tissue model parameters. Moreover, combinedwith this PVE correction, the CBF measurement method was used to acquire thedistribution image of CBF.Based on the CBF image, no significant difference of mean CBF was foundbetween the PTSD and non-PTSD groups. Considering the effect of age on theCBF, group comparisons of CBF were performed using analysis of two samplet-test in SPM8with age as covariates to account for the age effect. Comparisonsof CBF in survivors experienced mining disaster indicated that in right frontallobe, CBF of subjects with PTSD was obviously higher than those without PTSD.However, based on ROI analysis in patients with PTSD, no significant correlationof the CAPS score with mean CBF in right frontal lobe was observed.3. DTI analysisBased on the software of DTI studio and SPM, the DTI sequence wasanalyzed. Comparisons of FA and ADC in survivors experienced mining disaster indicated that in the right lateral ventricle, right fusiform gyrus, right superiortemporal gyrus, and right middle temporal gyrus, FA value of subjects with PTSDwas obviously lower than those without PTSD, and ADC value of subjects withPTSD was obviously higher in the right superior temporal gyrus and right corpuscallosum. However, based on ROI analysis in patients with PTSD, no significantcorrelation of the CAPS score with mean FA and/or ADC in above identifiedregions was observed. That may be due to subtle alteration of cortical thicknesscaused by early stage of PTSD.4. SummaryIn this study, the pipeline for cortical thickness measurement and analysiswas proposed, which improved computation speed and the accuracy of thicknessmeasurement. Meanwhile, considering the relatively low resolution and the effectof PVE on images, the PVE correction method based on MAP-EM algorithm wascombined with CBF measurement method to keep the accuracy of the CBF.These proposed methods may provide an efficient method for the corticalthickness and CBF analysis for nerve system or brain diseases.Based on proposed methods, in this study, a series of regions withsignificant difference between PTSD and non-PTSD groups were identified.Meanwhile, based on the software of DTI studio and SPM, the DTI sequence wasanalyzed. According to multiple regression analysis, the advantage ofmulti-modality image analysis is preliminarily confirmed. These results lay asolid foundation for the further study of pathological and etiological mechanism.Furthermore, the identified regions or indexes are hopeful to be the objective andquantitative indexes to diagnosis and evaluate PTSD in clinic.