KMID : 1103720070560020183
|
|
Journal of the Korean Society of Radiology 2007 Volume.56 No. 2 p.183 ~ p.189
|
|
Tuberculous Spondylitis vs Pyogenic Spondylitis: Focusing on the Discriminative MR Findings for Differentiation
|
|
Kim So-Young
Hong Suk-Joo Lee Chang-Yoon Jeong Kyu-Byung Park Chul-Min
|
|
Abstract
|
|
|
Purpose: The purpose of this study is to analyze the most discriminative MR findings for making the differential diagnosis of tuberculous and pyogenic spondylitis.
Materials and Methods: Sixty MR scans of 18 pyogenic spondylitis patients and 42 tuberculous spondylitis patients were retrospectively reviewed. The statistical analysis was performed using stepwise discriminant analysis and Student¡¯s T-test.
Results: The patients with tuberculous spondylitis had a significantly higher incidence of MRI findings as follows (p <0.05): smooth margin of a paraspinal mass/abscess in 67% [28/42] of the tuberculous spondylitis patients vs 6% [1/18] in the pyogenic spondylitis patients, mild endplate erosion in 52% [22/42] vs 38% [7/18], respectively, the presence of paraspinal mass/abscess in 100% [42/42] vs 6% [1/18], respectively, central dark signal intensity of the abscess in 86% [36/42] vs 39% [7/18]), respectively, subligamentous spread in 81% [34/42] vs 50% [9/18]), respectively, severe vertebral collapse in 20% [8/42] vs 11% [2/18]), respectively, and posterior extension in 62% [26/42] vs 33% [6/18]), respectively. Among of them, the significant discriminative MR findings were the margin of a paraspinal mass, the grade of endplate erosion and the presence or absence of a paraspinal mass in that order.
Conclusion: In the differentiation of tuberculous and pyogenic spondylitis, the margin of the paraspinal mass, the grade of endplate erosion and the presence or absence of a paraspinal mass are the most three discriminating MR findings in that order.
|
|
KEYWORD
|
|
Spine infection, Magnetic resonance(MR) comparative studies, Spondylitis
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|
|