Marieke van Rosmalen

Chapter 5 78 We obtained demographic and clinical data, including age, sex and time from onset of symptoms to research MRI (i.e. disease duration) in months. We performed a clinical examination and analyzed results from the diagnostic NCS of patients with CIDPandMMN to determine their clinical phenotype. We defined the following categories: sensorimotor, pure motor (i.e. no sensory involvement on clinical examination and NCS), and pure sensory or ataxic (i.e. no motor involvement on clinical examination and only demyelinating motor conduction abnormalities allowed). All participants gave informed consent. This study was approved by the Medical Ethical Committee of the UMCU. Equipment and MRI parameters We used a 3.0 Tesla MRI scanner (Philips Healthcare, Best, the Netherlands) with a 24-channel head- neck coil to image the cervical intraspinal roots. All participants were positioned in a supine position and scans were performed in a transversal slice orientation. We performed 3D balanced fast field echo with the following parameters: field of view = 250*250*39.90 mm 3 ; matrix size = 416*249; voxel size = 0.35*0.35*0.35 mm 3 ; echo time = 3.3 ms; number of echoes = 1; repetition time = 6.6 ms; flip angle = 55 o ; sense factor = 1.20 (P reduction foot/head) and 1 (S reduction anterior/posterior); acquisition time = 02:52 minutes. Additionally, we used coronal 3D TSE SPIR scans as anatomical reference, collected from a previously published MRI study. 11 The 3D TSE SPIR had the following parameters: field of view = 336*336*170 mm, matrix size = 224*223, voxel size = 0.75*0.75*1 mm 3 , echo time = 206 ms, repetition time = 2200 ms, turbo spin echo factor = 76, sense factor = 3 (P reduction right/left) and 1.5 (S reduction anterior/posterior), acquisition time = 03:59 minutes. Intraspinal root measurements on MRI We measured diameters of intraspinal nerve roots using Horos medical image viewer (version 3.3.6, www.horosproject.org ). To determine nerve root levels accurately we used reference lines in the coronal 3D TSE SPIR. We used zoom magnification (standardized to 250% for all images) and subsequently used the length tool to measure diameters (mm) of ventral and dorsal intraspinal nerve roots. Diameters were measured in the transversal plane perpendicular to the center lines of the nerve roots and in the middle of the intraspinal nerve root, bilaterally in nerve roots C5, C6 and C7 ( Figure 5.1 ), resulting in 12 measurements per subject. All measurements were performed thrice at each measurement site (approximately 7-10 minutes per subject) by one rater (MVR), who was blinded to the clinical status. We used the mean of these three measurements for further statistical analysis. We did not perform measurements when image quality was poor, when intraspinal nerve roots borders could not be accurately defined or when banding artifacts hindered an accurate measurement. Statistical analysis IBM SPSS Statistics (version 26, Chicago, Illinois, United States) was used for statistical analysis. To compare patient characteristics, we used a one-way ANOVA on numerical data, with post-hoc Tukey HSD to correct for multiple testing, and a Chi-squared test on categorical data. To evaluate the feasibility of our method we compared numbers of successfully performed measurements using