W Freund1, S Klessinger2,3, M Mueller4,5, M-E Halatsch3, G Hoepner4, F Weber6, B Schmitz7. 1. Klinik für Diagnostische und Interventionelle Radiologie, Universitätskliniken Ulm, Einsteinallee 23, 89081, Ulm, Deutschland. Freund-ulm@t-online.de. 2. Neurochirurgie, Nova Clinic Biberach, Biberach, Deutschland. 3. Neurochirurgie, Universitätskliniken Ulm, Ulm, Deutschland. 4. Klinik für Diagnostische und Interventionelle Radiologie, Universitätskliniken Ulm, Einsteinallee 23, 89081, Ulm, Deutschland. 5. Diagnostische und Interventionelle Neuroradiologie, Universitätskliniken Aachen, Aachen, Deutschland. 6. Neurologie, Bundeswehrkrankenhaus Ulm, Ulm, Deutschland. 7. Neuroradiologie, Universitätskliniken Ulm, Ulm, Deutschland.
Abstract
BACKGROUND AND OBJECTIVES: Angulated projections are standard in conventional radiography of the cervical spine, but rarely used in magnetic resonance imaging (MRI). As neuroforaminal pathology plays an important role in the etiology of radicular syndromes and may influence an operative approach, the utility of coronal oblique slices in MRI is explored. MATERIALS AND METHODS: In a retrospective setting, 25 consecutive patients with neurologically diagnosed cervical monoradiculopathy were identified. T2-weighted sagittal, coronal oblique, and transversal slice orientations were anonymized. Two radiologists and two neurosurgeons independently assessed the cases. Criteria were site, cause, and grading of the neuroforaminal stenosis and the level of confidence on a 100-point visual analog scale (VAS). We computed interrater agreement, sensitivity, and t tests. RESULTS: Using only one slice orientation, the sensitivity in detecting the relevant neuroforamen was 0.40 for transversal, 0.68 for sagittal, and 0.64 for coronal oblique scans. A combination of the different angulations increased sensitivity and in 4 cases only the coronal oblique scans proved diagnostic. The readers felt significantly more confident in attributing the cause of the pathology on coronal oblique planes (a mean of 72 VAS points, p = 0.0003 vs 58 (sagittal) vs 64 (transversal)). Interrater agreement was significantly better for experienced (kappa 0. 48) than for inexperienced readers (0.32, p = 0.02). CONCLUSIONS: Adding coronal oblique planes in cervical spine MRI increases sensitivity and confidence in attributing the cause of neuroforaminal pathology. They are regarded as useful by all the readers.
BACKGROUND AND OBJECTIVES: Angulated projections are standard in conventional radiography of the cervical spine, but rarely used in magnetic resonance imaging (MRI). As neuroforaminal pathology plays an important role in the etiology of radicular syndromes and may influence an operative approach, the utility of coronal oblique slices in MRI is explored. MATERIALS AND METHODS: In a retrospective setting, 25 consecutive patients with neurologically diagnosed cervical monoradiculopathy were identified. T2-weighted sagittal, coronal oblique, and transversal slice orientations were anonymized. Two radiologists and two neurosurgeons independently assessed the cases. Criteria were site, cause, and grading of the neuroforaminal stenosis and the level of confidence on a 100-point visual analog scale (VAS). We computed interrater agreement, sensitivity, and t tests. RESULTS: Using only one slice orientation, the sensitivity in detecting the relevant neuroforamen was 0.40 for transversal, 0.68 for sagittal, and 0.64 for coronal oblique scans. A combination of the different angulations increased sensitivity and in 4 cases only the coronal oblique scans proved diagnostic. The readers felt significantly more confident in attributing the cause of the pathology on coronal oblique planes (a mean of 72 VAS points, p = 0.0003 vs 58 (sagittal) vs 64 (transversal)). Interrater agreement was significantly better for experienced (kappa 0. 48) than for inexperienced readers (0.32, p = 0.02). CONCLUSIONS: Adding coronal oblique planes in cervical spine MRI increases sensitivity and confidence in attributing the cause of neuroforaminal pathology. They are regarded as useful by all the readers.
Authors: Anita R Gross; Jan L Hoving; Ted A Haines; Charles H Goldsmith; T Kay; Peter Aker; Gert Bronfort Journal: Spine (Phila Pa 1976) Date: 2004-07-15 Impact factor: 3.468