DTI for Parkinson's Disease: A Protocol for Systematic Review and Meta-Analysis

Background: There is no robust biological marker for the diagnosis of Parkinson's disease, and most of them are diagnosed until motor symptoms develop, which may affect the early intervention and prognosis Objective: To evaluate the diagnostic value of different parameter values of DTI in PD. Methods and analysis: We will systematically search the cochrane, pubmed, and embase databases, but we only consider observational studies and English studies. The main outcomes are DTI parameters, including FA, MDC, and ADC. We will evaluate the quality of the included studies through the NOS scale, and the data synthesis will be analyzed by revman5.3. Discussion: This systematic review will integrate all relevant DTI observation results on PD imaging, and evaluate whether DTI can be used as a biological marker for the diagnosis of PD. The review results will provide a useful reference for the diagnosis of PD Systematic review registration: The protocol has been registered at the International Platform of Registered Systematic Review and Meta-analysis Protocols(INPLASY202070098).


Background
Parkinson's disease (PD) is a neurodegenerative disease ,common in elderly people, insidious onset and slow progress. Its characteristic pathological changes are reduction of substantia nigra dopaminergic neurons and the formation of Lewy bodies ,leads to a decrease in dopamine transmitters in the striatum area.Resulting in clinical symptoms such as bradykinesia, resting tremor, muscle rigidity, and postural instability [1], accompanied by various non-motor symptoms [2], Such as olfactory dysfunction [3,4], cognitive and emotional disorders, sleep disorders, abnormal stools, gastrointestinal dysfunction [5,6], pain and sensory disturbances [7,8,9], etc. Seriously affect the patient's daily life, reduce life quality and loss work ability .
PD is the second largest neurodegenerative disease after Alzheimer's disease [10], about 1.2 million people in Europe suffer from PD [11], PD usually develops at the age of 65 years and older age,and overall incidence rate of 17 per 100,000 persons per year has been reported [12]. The incidence of PD has increased by more than double over the past 26 years, from 2.5 million patients in 1990 to 6.1 million patients in 2016 [13], the prevalence of PD will double from 6.9 million in 2015 to 14.2 million in 2040 [14]. Affected by the aging of the population, the PD population continues to increase, which will bring huge challenges in medical and socio-economic care [15].
PD is generally considered to be the result of the profound loss of dopamine (DA) neurons in the substantia nigra compacta (SNc) reaching the putamen [16] .Neurons in this area and other brain areas develop abnormal intracellular deposits known as Lewy bodies that contain aggregated α-synuclein [17],and motor symptoms associated with PD have been primarily attributed to this process [18].
The diagnosis of PD depends primarily on clinical signs and symptoms [19], although non-motor manifestations including depression, sleep problems and anosmia, typically begin years earlier, it is not diagnosed until the onset of motor symptoms [20]. In this context, reliable early prediction of disease would be essential for appropriate interventions and prognosis. Unfortunately, reliable biomarkers are still lacking. Diffuse MRI has been used to investigate the brain microstructural damage in PD patients, and it could predict the changes in bradykinesia and cognitive status over one year [21,22].A recent study have shown that DTI in early PD in many areas has a signi cant but subtle changes, such as the motor, premotor, and supplementary motor cortices, corpus callosum, and SN [23,24,25]. Previous studies that utilized DTI to evaluate the WM of PD patients with NCPs reported con icting observations, such as lower fractional anisotropy (FA) and higher mean diffusivity (MDC) [26,27,28,29] ,so we will carry out a systematic analysis to explore the diagnostic value of DTI for PD.

Protocol and registration
This systematic review protocol is being reported in accordance with the reporting guidance provided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) criteria (see Additional le 1) [30]. The review protocol has been registered within the International Platform of Exclusion criteria: Parkinson's disease, Parkinson's Overlay syndrome, case reports, submissions, qualitative studies, letters to the editor, comments, meeting minutes, reviews and meta-analyses.

Types of outcome measures
The primary outcome is fractional anisotropy (FA),secondary outcomes are mean diffusivity coeb cient (MDC) and apparent diffusion coeb cient(ADC).

Search methods for identi cation of studies
We will primarily search Cochrane , PubMed and EMBASE,studies regardless of publication date, but only english language will be included.Using the keywords "Parkinson's Disease","Diffusion tensor imaging "," fractional anisotrop(FA)"," Apparent diffusion coeb cient(ADC)","mean diffusion coe cient(MDC)" (see Additional le 2 for PubMed's search strategy).

Data extraction and management
Studies imported into EndNote V.X9 (Thompson Reuters, New York, New York, USA)after removing duplicates will be independently reviewed by two authors (XY,YL and ML) based on the exclusion and inclusion criteria. All data will be extracted and recorded independently by two reviewers (XYand YL) in an electronic database created in Microsoft O ce Excel V.2010.The information including the rst author's name, date of publication,type of study (cross-sectional, cohort / case-control studies), sample size, diagnostic criteria , mean and SD for FA,MDC,ADC score, 95% CI, and other relevant data for quality evaluation and risk of bias assessment.Any unclear information from an included article will be clari ed after contacting the corresponding authors and differences will be resolved by discussion with a third author.

Assessment of risk of bias in included studies
The Newcastle-Ottawa Scale will be used to evaluate the quality of studies. This scale is a nonrandomised controlled trial quality evaluation instrument with scores ranging from 0 to 9; scores of 0-4 and 5-9 mean low quality and high quality, respectively [31].

Data synthesis
The RevMan 5.3 software(The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) was used for data analysis. Both the χ 2 test and I² statistics will be used for the assessment of heterogeneity, and a xed effect model will be used if there is no obvious heterogeneity (I²<50% and p>0.1), with a random effects model being used if signi cant heterogeneity is found to exist (50% <I2<80% or p<0.01). For continuous data, we will calculate the mean difference (MD), and 95% con dence intervals (CI) if outcome measure scales are the same. In the case of different outcome measure scales, we will calculate the standardized mean difference (SMD) and 95% CI. If more than 10 articles are included, publication bias will be analysed by visual inspection of funnel plots.

Subgroup analysis
Subgroup analysis may eventually be carried out according to type of study (cross-sectiona /cohort/case report),if appropriate we will carried out subgroup analysis according to different areas of the brain

Sensitivity analysis
If a meta-analysis is performed, a sensitivity analysis will be conducted, excluding studies from the analysis one by one. These will be performed to examine the potential in uence of each study in the pooled estimates.

Con dence in cumulative evidence assessment
The review will evaluate the con dence of the body of cumulative evidence using the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) [32] approach. We will assess the strength of evidence using ve criteria: risk of bias, inconsistency, indirectness, imprecision, and publication bias. We will rate the overall level of certainty for each outcome as high, moderate, low, or very low. Results will be presented in tables for each primary outcome.

Discussion
This review will synthesize the parameter of DTI on PD's imaging observations, summarize whether different parameter values are consistent in PD patients, and whether they can be used as biological markers for PD diagnosis and prediction. It will also help discover the gaps in existing studies, provide new directions for future research and nd new biological markers for PD diagnosis. Declarations DY ,XC, and ML designed the study. XY, ML, SJ, and YL drafted the manuscript, and DYand XCinitiated the study design. XY and YL developed the search strategy. All authors contributed to the re nement of the study protocol, reviewed, and provided feedback on the manuscript and approved the nal manuscript. DY and XC serves as the guarantor of the manuscript. The authors read and approved the nal manuscript.

Funding
This work was supported by an National key research and development Program(No.2018YFC1705604).
Availability of data and materials Not applicable.
Ethics approval and consent to participate Systematic review-not applicable.

Consent for publication
Not applicable.