Quantitative Evaluation Of Cerebrospine Fluid Hydrodynamics In Thoracic Spine With3.0T MR Fast Cine-PC

PartⅠ The experimental study of fluid quantitative measurement using3.0TMRI fast cine phase contrast sequenceObjective:To evaluate the feasibility and accuracy of MR fast cinephase-contrast technique in fluid flow quantification by establishing flowphantom.Methods:The phantom consisted of a plastic tube with diameter of4.75mm,a high pressure injector,0.9%physiological saline.0.9%physiological saline was injected into the plastic tube, and fast cine phasecontrast sequence was performed by using3.0T MRI.The scanning at differentflow rates(0.1ml/s,0.2ml/s,0.3ml/s,0.4ml/s,0.5ml/s,0.6ml/s,0.7ml/s,0.8ml/s,0.9ml/s), different velocity encoding(5cm/s,10cm/s,15cm/s,20cm/s,25cm/s,30cm/s,35cm/s,40cm/s),different directions of flow(caudal–cranial, cranial-caudal), different vertical angles(200,300,450), horizontalangles(200,300,450)was measured.After postprocessing, measured values werecompared with actual values by satistical analysis.Results:There was no statistical difference between measured values andactual values at different flow rate(0.1ml/s,0.2ml/s,0.3ml/s,0.4ml/s,0.5ml/s,0.6ml/s,0.7ml/s,0.8ml/s,0.9ml/s)(t=1.072,p=0.315); The aliasingartifacts appeared at the velocity encoding of5cm/s.In other velocityencoding aliasing artifacts did not appear. There was no statistical differencebetween measured values and actual values at different velocity encoding(t=0.351,p=0.739)；No statistical difference between Measured values andactual values(t=1.814,p=0.121)at different angle, measured values and0angle(t=0.456, p=0.673).Conclusions:3.0T MRI fast cine phase contrast sequence allows accuratemeasurement of flow velocity with fluid flow phantom, and provides reliable experimental evidence for the quantitative study of anybfluid flow.PartⅡ Optimal Parameters for Imaging Cerebrospine Fluid Flow in thoracicspine with3.0T MR Fast Cine phase contrast sequenceObjective: To optimize imaging parameters of cerebrospine fluid flow inthoracic spine with3.0T MR fast cine phase contrast.Methods:30healthy individuals (12men,18women) aged between21and55years(mean36.16±13.42years) were enrolled for the MRImeasurement of spine cerebrospine fluid flow. Fast cine phase contrastsequence was performed located at T6-7Level. The imaging effect wasexplored with different parameters, including velocity encoding(5cm/s,10cm/s,15cm/s,20cm/s),frequency encoding direction(R/L, A/P), FOV(18cm,26cm,34cm), matrix(384×256,256×192)and NEX(1,2),respectively.Results: The number of excellent, good and poor imaging in the velocitycodes of5cm/s,10cm/s,15cm/s,20cm/s were0,14,16;19,11,0;28,2,0;30,0,0; respectively(χ2=8.65,p=0.02). The number of excellent, good andpoor imaging with different FOV(18cm,26cm,34cm) were0,18,12;11,18,1and29,1,0(χ2=8.77,p=0.01). The number of excellent and poorimaging with different frequency encoding direction(R/L, A/P) were27,3and10,20, respectively(Z=－4.12,p＜0.001). The number of excellent andgood imaging with different matrix(384×256,256×192)were29,1and3,27(Z=－4.81,p＜0.001); with different NEX(1time,2times)were6,24and30,0, respectively(Z=－4.89,p＜0.001). The optimal parameters wereobtained and they were velocity encoding10cm/s, frequency encodingdirection A/P, FOV26cm, matrix256×192and NEX2times,respectively.Conclusions: The better images and accurate quantitative assessment ofthoracic spine cerebrospine fluid flow are performed by optimizing theparameters of3.0T MR fast cine phase contrast sequence.PartⅢ Quantitatively evaluate cerebrospine fluid flow in the thoracic spineObjectives: The aim of this study was to quantitatively evaluatecerebrospine fluid flow in the thoracic spine by using fast cine phase-contrast magnetic resonance imaging (MRI) technique,and to explore the influencefactor, flow characteristics of cerebrospine fluid.Methods:92healthy individuals (40men,52women) aged between21and72years(mean45.09±14.79years) were enrolled for the MRImeasurement of thoracic spine cerebrospine fluid flow. Fast cinephase-contrast sequences were acquired at the level of T2-3, T4-5, T6-7, T8-9,T10-11. At these levels, caudal peak flow velocity,cranial peak flow velocity,caudal peak volume flow,cranial peak volume flow,caudal mean flowvelocity,cranial mean flow velocity,caudal mean volume flow,cranial meanvolume flow were studied.Subjects were divided into five age groups:20-29years(18cases),30-39years(18cases),40-49years(17cases),50-59years(19cases), and≥60years(20cases).The volunteers were dividedinto male and female groups. The values of cerebrospine fluid flow among allage groups, between male group and female group, among the level of T2-3,T4-5, T6-7, T8-9, T10-11, between cranial and caudal directions werecompared,respectively.We observed the correlation between the thoracicspine cerebrospine fluid flow and the corresponding level of the spine cordarea, spine anteroposterior diameter, subarachnoid anteroposteriordiameter.The thoracic spine subarachnoid space was artificially divided intofour parts: Ventral, dorsal,left, right and they were compared. The revelanceof cerebrospine fluid flow in four parts and subarachnoid space,spineanteroposterior diameter were studied.Results: There was a statistical significance in cerebrospine fluid caudalpeak flow velocity between the age group of≥60years and the other threeage group(sp=0.029,p=0.022,p=0.020). Cranial peak flow velocity of≥60years was lower than that of the other agegroups(p=0.009,p=0.016,p=0.014,p=0.012). In≥60years group caudal peakvolume flow,cranial peak volume flow,caudal mean flow velocity,cranialmean flow velocity of cerebrospine fluid were lower than in the other agegroups,but there was no statistical significance(p>0.05). Statistical differenceswas not detected in flow parameters between male and female(p>0.05). There was a statistical difference in cerebrospine fluid caudal peak flow velocity,caudal peak volume flow, caudal mean flow velocity, caudal mean volumeflow between the level of T10-11and T2-3,T4-5,T6-7,p volues were0.007,0.008,0.049；0.004,0.006,0.025；0.003,0.004,0.026；0.001,0.005,0.041,respectively. There was no statistical difference in cerebrospine fluidcranial peak flow velocity, cranial peak volume flow, cranial mean flowvelocity, cranial mean volume flow between the level of T10-11andT2-3,T4-5,T6-7(p>0.05).Cerebrospine fluid caudal flow was higher thancranial flow in all individuals. Caudal and cranial mean volume flow showed apositive correlation with subarachnoid space area. There was a positivecorrelation between cranial peak flow velocity,cranial peak volume flow andspine cord anteroposterior diameter, a negative correlation between caudalpeak volume flow and spine cord anteroposterior diameter.In segmentmeasurement, caudal and cranial mean flow velocity of ventral was the largest,dorsal was the lowest.No statistical difference was detected between left andright(p>0.05). There was a positive correlation between caudal peak flowvelocity,volume flow of ventral and subarachnoid space anterior diameter, anegative correlation with posterior diameter, but dorsal was reverse.Conclusions: Thoracic spine cerebrospine fluid flow of normal adultcan be quantitatively evaluated by MRI,and age is a effective factor:cerebrospine fluid flow in≥60years is lower than the other aged groups.Thoracic spine CSF flow is not affected by gender. Thoracic spinecerebrospine fluid flow in below level of intervertebral disc is lower. Caudalcerebrospine fluid flow is higher than cranial.There is a difference betweenfour segments of thoracic spine canal; ventral is the largest, dorsal is thelowest,left and right sides are stable. Thoracic spine cerebrospine fluid flow isaffected by the spine subarachnoid space area and the anteroposteriordiameter.PartⅣ The influence of respiration on cerebrospine fluid flow in thoracicspineObjectives:To quantitatively evaluate the changes of cerebrospine fluid flow velocity and volume flow of thoracic spine and the total cerebral bloodflow in breath-holding after inspiration or expiration.To observe thecorrelation of cerebrospine fluid flow velocity and volume flow of thoracicspine and the total cerebral blood flow.Methods:32healthy individuals (15men,17women) aged between23and65years(mean41.38±12.74years) were enrolled for the MRImeasurement of spine CSF flow. Sagittal T2-weighted FSE images in thoracicspine and cervical spine were acquired first. To quantify bilateral internalcarotid artery and vertebral artery flow,fast cine-phase contrast sequence wasacquired located in C2-3level under free breathing, breath-holding afterinspiration, breath-holding after expiration. Flow velocity and volume flow ofeach artery among three states of respiration status werecompared,respectively.We evaluated the agreement between bilateral internalcarotid artery flow, bilateral vertebral artery flow,and calculated the totalcerebral blood flow,respectively.We measured the cerebrospine fluid flowlocated in T6-7level under free breathing, breath-holding after inspiration,breath-holding after expiration and obtained the caudal peak flow velocity,cranial peak flow velocity,caudal peak volume flow,cranial peak volumeflow,caudal mean flow velocity,cranial mean flow velocity,caudal meanvolume flow,cranial mean volume flow in a cardiac cycle. The relevance offlow velocity and volume flow of cerebrospine fluid and each artery werestudied.Results: In bilateral internal carotid artery and vertebral artery flowvelocity and volume flow under breath-holding after inspiration, breath-holding after expiration were significantly higher than that under freebreathing(p<0.05).There was no difference among three respiration states ofbilateral internal carotid artery mean volume flow and bilateral vertebral arterymean volume flow(p>0.05). The total cerebral blood flow under freebreathing, breath-holding after inspiration, breath-holding after expiration was797.89ml/min,972.15ml/min,985.96ml/min,respectively. The total cerebralblood flow under breath-holding after inspiration, breath-holding after expiration was significant higher than free breathing(p<0.05). Flow velocityand volume flow of cerebrospine fluid under breath-holding after inspiration,breath-holding after expiration were significantly higher than that under freebreathing(p<0.05),but no statistical significance in cranial peak flowvelocity, cranial peak volume flow(p>0.05). There was the correlationbetween total cerebral blood flow and cerebrospine fluid in the mean volumeflow of breath-holding after inspiration and free breathing (r=0.504,p=0.039),breath-holding after expiration and free breathing(r=0.389,p=0.042).Conclusions:1) Flow velocity and volume flow in bilateral internalcarotid artery and vertebral artery under breath-holding after inspiration,breath-holding after expiration are significantly higher than that under freebreathing, which demonstrates that the total cerebral blood flow increases;2)Flow velocity and volume flow of cerebrospine fluid under breath-holdingafter inspiration, breath-holding after expiration are significantly higher thanthat under free breathing;3) There is correlation between total cerebral bloodflow and cerebrospine fluid in the mean volume flow.PartⅤ The effect of posture on cerebrospine fluid flow in thoracic spineObjective: To investigate the influence of posture on cerebrospine fluidflow in thoracic spine by using fast cine phase contrast MRI.Methods:31healthy individuals (13men,18women) aged between21and65years(mean39.5±12.6years) were enrolled for the MRI measurementof spine CSF flow. Sagittal T2-weighted FSE images of Supine position inthoracic spine were acquired first. Fast cine-phase contrast sequence wasscanned located in T6-7level. After supine scanning, rescanning wasperformed with cummerbund.After five minutes, the prone position imagingwere acquired located in T6-7level. We obtained the values of cerebrospinefluid flow with different posture: caudal peak flow velocity, cranial peak flowvelocity, caudal peak volume flow,cranial peak volume flow,caudal meanflow velocity,cranial mean flow velocity,caudal mean volume flow,cranialmean volume flow. Cerebrospine fluid flow velocity and volume flowbetween supine position and prone position, supine position and cummerbund were compared.Results: Caudal peak flow velocity, caudal peak volume flow, caudalmean flow velocity, caudal mean volume flow in supine position were largerthan in prone position,and there was statistical significance between twoformers(p<0.05). There were no statistical significance between supineposition and prone position in cranial peak flow velocity, cranial peak volumeflow, cranial mean flow velocity, cranial mean volume flow.(p>0.05). Thedifference was not detected between cummerbund and supine position incaudal peak flow velocity, cranial peak flow velocity, caudal peak volumeflow,cranial peak volume flow,caudal mean flow velocity,cranial mean flowvelocity,caudal mean volume flow,cranial mean volume flow.(p>0.05).Conclusion: The effect of posture on cerebrospine fluid flow in thoracicspine can be quantified by MRI. Caudal cerebrospine fluid flow velocity andvolume flow in supine position are larger than in prone position. Howeverthere is no relation with cummerbund.PartⅥ Characteristics of cerebrospine fluid flow of thoracic spine in patientswith cervical spondylosisObjective: To quantitatively evaluate cerebrospine fluid flow velocityand flow volume of thoracic spine in patients with cervical spondylosis and toobserve the correlation between the severity of cervical spondylosis andcerebrospine fluid flow velocity and flow volume of thoracic spine.Methods: A total of91patients with cervical spondylosis were studied,40men and51women, and the mean age was45.21±14.58years (range,28–72years). All patients underwent fast cine phase contrast MRI at T6-7level. The degree of cervical spondylosis was rated as low-, intermediate-, orhigh-grade. Low-grade was defined as involving no effacement of thesubarachnoid space, intermediate-grade as involving effacement of this space,and high-grade as involving effacement of this space, together withcompressive myelopathy.23cases of Low grade, intermediate-grade30cases,high-grade38cases.Flow velocity and flow volume were compared betweenpatient of cervical spondylosis and normal volunteers. Clinical scoring system used the Japanese Orthopedic Association (JOA) score.The correction of JOAscore and The severity of cervical spondylosis were evaluated. According tothe increased signal intensity (ISI) on T2WI, the patients of high-grade weredivided into the group with or without ISI and correlated them in cerebrospinefluid flow velocity and flow volume of thoracic spine.Results: There was no difference between low-grade patients and normalvolunteers(p>0.05); Significant difference not only betweenintermediate-grade patients and normal volunteers (p<0.05),but also betweenwas high-grade patients and normal volunteers. There was significantdifference in caudal peak flow velocity,cranial peak flow velocity,caudalpeak volume flow,cranial peak volume flow,caudal mean flow velocity,cranial mean flow velocity,caudal mean volume flow,cranial mean volumeflow among the three groups, high-Grade patients had a lower cerebrospinefluid flow velocity and flow volume of thoracic spine(p<0.05).The mean JOAscore in patients of low-, intermediate-, and high-grade were13.35±1.36,12.41±1.65,10.14±1.95, respectively. A high correlation between theJapanese Orthopedic Association score and the cerebrospine fluid flowvelocity and flow volume in different degree patients was demonstrated.Cerebrospine fluid flow velocity and flow volume of high-grade patients withISI were greater,but there was no statistical significance.Conclusions:1)According to presence or absence of subarachnoid andspine cord compression or not on the sagittal T2images, the degree of cervicalspondylosis is rated,which is helpful for clinicians.2)The relevance of theseverity degree in cervical spondylosis and the Japanese OrthopedicAssociation score, cerebrospine fluid flow velocity and flow volume ofthoracic spine are showed.3) There is no statistical difference betweenhigh-grade patients with and without ISI.