Devoted to diagnostic and interventional spine imaging and therapeutics


Reliability of Readings of Magnetic Resonance Imaging Grades of Cervical Neural Foraminal Stenosis: Comparison of 2D Conventional and 3D Isotropic Protocol 2013

Category General Spine Chang Ho Kang
Kyung-Sik Ahn
Woo Young Kang
Jae Shik Shin
Yu-Whan Oh
Jeongrim Lee
Purpose To determine and compare the reliability of 2D and 3D MRI interpretations in cervical neural foraminal stenosis Materials & Methods Forty-one patients with possible cervical spinal stenosis prospectively underwent cervical spine MRI in which the protocol included 2D T1 and T2-weighted fast spin-echo sequences in sagittal and axial planes, and a single sagittal 3D T2-weighted fast spin-echo sequence. The sagittal 3D data sets were reformatted at 1-mm slice thickness without a gap in 2 oblique sagittal plane (perpendicular to both right and left neural foramen). Three radiologists independently and blindly rated the severity of neural foraminal stenosis at C2-3 to C7-T1 with a 4-point scale, which was done in two separate 2D and 3D interpretation sessions with more than a month of interval. Kendall’s rank correlation test was used to characterize inter-reader reliability for categorical rating data. Results Overall, interobserver agreement in ordinal ratings of neural foraminal stenosis was substantial in 2D (Kendall tau-b, 0.61-0.89) and substantial to excellent in 3D MRI (Kendall tau-b, 0.61-0.92). Interobserver agreement at C3-4 to C6-7 level was slightly higher in 3D protocol than in 2D protocol (Kendall tau-b: C3-4, 0.80 vs. 0.78; C4-5, 0.89 vs. 0.78; C5-6, 0.90 vs. 0.88; C6-7, 0.90 vs. 0.84). At C2-3 and C7-T1, reliability of reading of cervical neural foraminal stenosis did not differ between the two MR protocols (Kendall tau-b: C2-3, 0.61; C7-T1, 0.71). Conclusion We found that the MRI reading of cervical neural foraminal stenosis showed substantial interobserver agreement or above, and that 3D isotropic protocol slightly reduced the degree of interobserver variability. References 1. Modic MT, Masaryk TJ, Ross JS, et al. Cervical radiculopathy: value of oblique MR imaging. Radiology. 1987;163:227-31. 2. Roberts CC, McDaniel NT, Krupinski EA, et al. Oblique reformation in cervical spine computed tomography: a new look at an old friend. Spine. 2003;28:167-70. 3. Goodman BS, Geffen JF, Mallempati S, et al. MRI images at a 45-degree angle through the cervical neural foramina: a technique for improved visualization. Pain Physician. 2006;9:327-32. 4. Kuijper B, Beelen A, van der Kallen BF, et al. Interobserver agreement on MRI evaluation of patients with cervical radiculopathy. Clin Radiol. 2011;66:25-9. 5. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159-74.