Clinical validity of two different grading systems for lumbar central canal stenosis: Schizas and Lee classification systems.

OBJECTIVE: To assess the learnability of two magnetic resonance imaging (MRI) grading systems for lumbar central canal stenosis based on inter-observer agreement and test-retest reliability of doctors with no prior knowledge of the two systems. MATERIALS AND METHODS: Two clinical fellows, one novice radiology resident, one neurosurgeon, and one orthopedic surgeon, who were unaware of the two qualitative MRI grading systems prior to this study, acquainted themselves with the teaching files. All five observers independently assessed the LCCS grade of 70 patients using T2-weighted axial magnetic resonance images at the L2-3, L3-4, L3-4, and L5-S1 disc levels. Analysis was performed twice at an interval of two months. RESULTS: The inter-observer agreement among all five readers was excellent and test-retest reliability was moderate to excellent for both the Schizas and Lee systems. Positive percentage agreements were found to be over 0.8 in almost all observers with relatively narrow 95% confidence limits. CONCLUSION: Both Schizas and Lee MRI grading systems for LCCS are reliable grading systems, and can be used as a learnable method for both clinicians and radiologists.

Comparison of global left ventricular function using 20 phases with 10-phase reconstructions in multidetector-row computed tomography.

To compare the measurement of global left-ventricular (LV) function parameters of 64-slice multidetector-row computed tomography (MDCT) between 20- and 10-reconstruction phases. Fifty five patients with suspected or known coronary artery disease underwent 64-slice MDCT. LV end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), and ejection fraction (EF) were measured from MDCT data sets using threshold-based volume segmentation and reconstruction at every 5% (20 phases) and 10% (10 phases) step through the R-R interval. These global functional parameters were compared to those obtained via two-dimensional transthoracic echocardiography (2D-TTE), considering the reference standard. The required time for CT data analysis was checked. Agreement for parameters of LV global function was determined using Pearson's correlation coefficient (r) and Bland-Altman analysis. LV volumes (EDV(-5%) 87.5 ± 17.1 ml, EDV(-10%) 87.7 ± 16.3 ml; ESV(-5%) 32.4 ± 10.6 ml, ESV(-10%) 31.9 ± 9.9 ml; SV(-5%) 55.1 ± 10.5 ml, SV(-10%) 55.8 ± 9.9 ml; mean ± SD) and EF (EF(-5%) 63.4 ± 6.2%, EF(-10%) 63.9 ± 5.8%) did not differ significantly between the 20- and 10 phase reconstructions, and evidenced good to excellent correlation (r = 0.786-0.896, all P < 0.001) with the 2D-TTE results. The mean required time for CT data analysis in the 20- and 10 phase reconstructions were 15.5 ± 4.0 and 7.3 ± 2.5 min. Within MDCT, using 10-phase image reconstruction is sufficient to evaluate LV volumes and EF, and is also more time-effective than 20-phase reconstruction.