Novel Magnetic Resonance Imaging In Diagnosis Of Parkinson’s Disease

Background and Purpose:Parkinson’s disease(PD)is a second common neurodegenerative disease in the elderly people and its morbidity only after Alzheimer’s disease.The loss of dopamine neurons and neuromelanin in substantia nigra play an important role in pathological change of PD.At present,lacking of imaging indicators that can objectively and accurately detect the pathological changes of substantia nigra in PD,diagnosis of PD still reliant on clinical symptoms such as resting tremor,bradykinesia,rigidity,postural instability,and gait impairment,but the misdiagnosis rate of clinicians for PD is as high as 20-30%.Early diagnosis and accurate evaluation of PD are very important for effective treatment,modifying treatment plan and improvement life quality of patients.However,conventional magnetic resonance imaging(MRI)can provide limited information and cannot effectively distinguish PD patients from the healthy.Therefore,exploring new techniques that can accurately detecting the pathological changes of substantia nigra in PD has been the focus and direction of research worldwide.Recent years,MRI technique has developed rapidly,and some novel MRI techniques have emerged.These advanced MRI techniques can provide more information reflecting tissue characteristics than conventional MRI,which provides the possibility for non-invasive detection of PD nigra pathological changes.Magnetic resonance fingerprint imaging(MRF),effective transverse relaxation rate(R2*),magnetic sensitivity weighted imaging(SWI)and quantitative magnetic susceptibility map(QSM)are novel MRI techniques developing in recent years,which can reflect the relaxation time and iron composition changes of substantia nigra in PD,respectively,and can provide more effective information than conventional MRI.It has great potential in accurate diagnosis of PD and evaluation of disease progression.In this study,new MRI techniques including MRF and R2*,SWI and QSM generated by Strategically Acquired Gradient Echo(STAGE)were used to study the pathological changes of substantia nigra in patients with PD,exploring new imaging markers for accurate diagnosis of PD by detecting the pathological changes of substantia nigra in PD.Materials and methods:This study included 40 PD patients(26 males,14 females,age range:51-70 years,mean age:61 years)and 40 age-and sex-matched healthy controls(26 males,14 females,age range:51-69,mean age:61 years)in our local hospital from October 2017 to December 2018.All patients and healthy controls underwent MRF scans on a 3T high-field magnetic resonance imager(Magnetom Skyra,Siemens Healthcare,Erlangen,Germany),which was equipped with a 20-channel head/neck coil.MRF generates quantitative T1 and T2 map simultaneously by matching the collected signal with a dictionary.All patients and healthy controls performed STAGE scanning on a 3.0T high-field magnetic resonance imager(Magnetom Prisma,Siemens Healthcare,Erlangen,Germany),which was equipped with 64-channel head coils.The STAGE sequence needs to be scanned twice,and the location of the two sancsi sidentical,the location line is parallel to the anterior and posterior corpus callosum line,and covers the whole brain.Image analysis was carried out after data acquisition and reconstruction.First of all,Regions of interest(RO1)was drawn on the T2 map generated by MRF and calculated T2 value of the both side of substantia nigra with the largest display layer of the substantia nigra.Then,copy the ROI of the both side of substantia nigra to T1 map,R2*map and QSM map,respectively.Analysis of "swallow tail sign":according to the double-blind principle,the existence of "swallow tail sign"(the high signal band composed of nigrosome-1 in the posterior part of substantia nigra)on SWI was evaluated by two radiologists with 5 years and 15 years radiological experience of nervous system diseases,respectively.The images were divided into abnormal group(one side "swallow tail " disappeared or both disappeared)and normal group(both sides"swallow tail" existed,or one side existed,the other side may exist).Statistical analysisAccording to the distribution of data,x 2 or Fisher exact test was used to compare qualitative parameters,and t test or Wilcoxon rank sum test was used to analyze continuous variables.Then the imaging indexes with statistical differences were analyzed by Logistic regression to evaluate the diagnostic efficiency of the combined indexes.Then Spearman correlation analysis was made between the imaging indexes of T1,T2,R2*and QSM and the course of PD,the motor symptom score of the third part of UPDRS and the severity of the disease obtained by evaluating Hoehn-Yahr stage.Then the subject working characteristic curve and the area under the curve(area under the curve,AUC)of continuous variables were drawn to evaluate the diagnostic performance of T1,T2,R2*and QSM values.The best offcut value was determined according to the Youden index,and the sensitivity and specificity of the best offcut value were calculated,and the AUC,sensitivity and specificity of each parameter were statistically analyzed to determine the best index for the diagnosis of PD.For each variable,kappa test was used to check the consistency between observers.When P<0.05,the difference was considered as statistically significant.Results:1.The T1 values of left and right substantia nigra in PD patients(mean±tstandard deviation,left substantia nigra:992.77±28.09ms;right substantia nigra:991.05±33.30ms)were significantly higher than those in normal control group(left substantia nigra:948.36±30.04ms;right substantia nigra:954.224±31.39ms,P<0.001).The R2*values of left and right substantia nigra in patients with PD(mean±standard deviation,left substantia nigra:41.37±11.80ms;,right substantia nigra:36.38±7.74ms)were significantly higher than those in normal controls(left substantia nigra:31.358.38ms;,right substantia nigra:29.87±6.16 ms,P<0.05).The QSM values of left and right substantia nigra in patients with PD(mean±standard deviation,left substantia nigra:0.183±0.038 ms;,right substantia nigra:0.169±0.029 ms)were significantly higher than those in normal controls(left substantia nigra:0.136±0.034 ms;,right substantia nigra:0.131±0.030 ms,P<0.001).2.The T1 value of the left substantia nigra was correlated with the severity of the disease and the score of motor symptoms(r=0.803 and 0.438,P<0.001 and 0.005,respectively),but not with the course of the disease(P=0.122).The T1 value of the right substantia nigra was not correlated with the severity of the disease,the score of motor symptoms and the course of disease(P=0.557,0.464 and 0.282,respectively).The left substantia nigra of PD was weakly correlated with the score of motor symptoms(r=0.403,P=0.010),but not with the severity of the disease and the course of disease(P=0.709 and 0.505,respectively),and the value of R2*of the right substantia nigra was not correlated with the severity of the disease,the score of motor symptoms and the course of disease(P=0.539,0.363 and0.612,respectively).The QSM of the left substantia nigra was correlated with the severity of the disease and the score of motor symptoms(r=0.318 and 0.441,P=0.046 and 0.004,respectively),but not with the course of the disease(P=0.606),and the QSM of the right substantia nigra was not correlated with the severity of the disease,the score of motor symptoms and the course of the disease(P=0.182,0.220 and 0.144,respectively).3.The results of correlation analysis showed that the T1 value of the left substantia nigra was weakly correlated with the left T2,R2*and QSM(r=0.337,0.485 and 0.315,P=0.033,0.002 and 0.047,respectively),while the T1 value of the right substantia nigra was weakly correlated with the right T2 and QSM(r=0.491,0.367,P=0.001 and 0.020),but not with the right R2*(P=0.248).In patients with PD,the left T2 value was weakly correlated with the left R2*(r=0.395,P=0.012),moderately correlated with the left QSM(r=0.526,P<0.001),and the T2 value of the right substantia nigra was weakly correlated with the right R2*and QSM(r=0.393,0.319,P=0.012,0.045).There was a strong correlation between the R2*value of the left substantia nigra and the right QSM in PD(r=0.736,P<0.001),and a moderate correlation between the R2*value of the right substantia nigra and the right QSM(r=0.508,P<0.01).Details are shown in Table 2.4.The results of "swallow tail sign" analysis showed that the accuracy,sensitivity,specificity,positive predictive value and negative predictive value of PD and healthy control were 86.25%,87.5%,85%,85.37%and 87.18%,respectively.5.Further multivariate regression analysis of T1,QSM,R2*and "swallow tail sign" of substantia nigra showed that T1 value,QSM value and swallow tail sign of substantia nigra were important predictive factors for the diagnosis of PD.When the above three indexes were combined,AUC reached the highest value of 0.968,the sensitivity was 92.5%,and the specificity was 90%.In addition,the ROC results showed that the AUC of T1 was 0.927,the best cutoff value was 971.15ms,the corresponding sensitivity was 90%and the specificity was 90%.The AUC of QSM is 0.897,and the best cutoff value is 0.15 ppb with sensitivity of 82.5%and specificity of 85%,respectively.The AUC of the left T1 value is 0.881,and the best cutoff value is 973.5 ms,with sensitivity of 82.5%and specificity of 85%.The AUC of the right T1 value is 0.811,and the best cutoff value is 963.9 ms with a sensitivity of 87.5%and specificity of 72.5%.Conclusions:1.In the differentiation of Parkinson’s disease and healthy control,the diagnostic efficiency of T1 map generated by MRF is better than that of R2*and QSM,and the diagnostic efficiency of T1 map combined with "swallow tail sign" and QSM is best in the differentiation of Parkinson’s disease and healthy control.2.The quantitative parameters of MRF and STAGE presented a positive correlation with clinical information of PD patients,such as the score of motor symptoms and the disease severity.Therefore,MRF and stage can quantitatively detect the disease development of PD patients.3.Noninvasive MRF and STAGE provide new imaging markers for accurate diagnosis of PD and assessment of disease progression.