Contribution of the mesencephalon indices to differential diagnosis of parkinsonian disorders

Parkinson's disease (PD) is a progressive, chronic disease characterized by cardinal parkinsonism clinical signs and it usually begins at the age of 50-60. Atypical parkinsonian disorders (APD) such as Progressive Supranuclear Palsy (PSP) and Multiple System Atrophy of the Parkinson type (MSA-P) are diseases that have additional characteristic clinical signs of cardinal parkinsonism and have poor prognosis than PD. An early distinction between atypical parkinsonian disorders and PD is important because of differences in prognosis and treatment response. There is no biological marker available for this distinction and the clinical data are determinative in the final diagnosis. However, the average time from symptom onset to diagnosis is 3.6 to 4.9 years. In this process, patients with APD group diseases with different prognosis and treatment, such as MSA-P and PSP, can be followed up for many years by misdiagnoses such as idiopathic Parkinson's disease, balance-movement disorder ,cerebrovascular disease and depression etc. until the characteristic clinical characteristics are established. Postmortem series and MRI studies suggest thar, superior cerebellar peduncles and midbrain atrophy are present in PSP, middle cerebellar peduncles and pons atrophy are present in MSA-P. By contrast, all these brain structures are spared in PD. Thus, the purpose of our study was to assess the sensitivity and specificity of brainstem structures' morphologic changes for differentiating PSP from PD and the MSA-P, by single and combined MR imaging measurements. From November 2015 to May 2017, MRI of 14 patients with possible PSP, 43 patients with PD, 8 patients with probable MSA-P, and 45 age-matched controls were recruited in this retrospective study. Clinical diagnoses was made according to established criteria by a neurologist who is experienced in movement disorders more than ten years. The pons area (P), mesencephalon area (M), middle peduncle width (MCP), superior peduncle width (SCP), and peduncle angle (PA) were measured from T1 MP-RAGE images. In addition to this the P/M ratio, MCP/SCP ratio, the previously defined MR parkinsonism index [(P/M)x(MCP/SCP)] and also index that we termed the Akdeniz index was calculated [(P/M)x(PA/180)]. Two independent radiologist' (one of them was an experienced neuroradiologist) who were blinded to the patients' diagnoses evaluated all MR images, thus interrater reliability was established. As well, a second evaluation was made 2 weeks after by one of researchers to assess the intrarater reliability. There was a statistically significant difference among the PSP, MSA and PD patients' measurements. It is understood after the pairwise comparison that the significantly is due to the difference between PSP patients and PD patients. M and SCP were significantly smaller in PSP patients compared to PD patients. P/M ratio, MCP/SCP ratio and PA were significantly larger in patients with PSP. MR Parkinsonism Index (PI) ratios were significantly higher in the PSP (median, 16.56) group than in the PH (median, 10.9) and MSA (median, 8.37) groups. Likewise, Akdeniz İndex (AKI) ratios are significantly higher in the PSP (median, 2.63) group than in the PH (median, 1.68) and MSA (median, 1.48) groups. When PI and AKI were compared, similar sensitivity and specificity values were obtained. However, inter- and intrarater reliability were significantly higher in AKI compared to PI. AKI and PI can help distinguish patients with PSP from those with PD. However, because of the easier measurement and the higher agreement between measurements, we believe that the use of AKI is preferable. Key Words: MRG, Parkinson disease, Progressive Supranuclear Palsy, Multiple System Atrophy of the Parkinson type, Parkinson index