Abstract
PURPOSE: To investigate the diagnostic efficacy of a new four-area method for evaluating biochemical changes in lumbar facet joint (LFJ) degeneration on T2* mapping. METHODS: LFJ degeneration was morphologically graded on T2-weighted imaging using the Weishaupt system. T2* value of LFJ was measured on T2* mapping using both all-inclusive and four-area methods. Inter-observer reliability for continuous and categorical variables was evaluated using intraclass correlation coefficient (ICC) and Kappa value, respectively. The correlation between continuous variables and ordered categorical variables was examined using one way ANOVA or Kruskal-Wallis test, as appropriate. RESULTS: Fifty-eight patients with LBP underwent standard MRI protocols and axial T2* mapping. In all-inclusive method analysis, the median T2* value of grade 0 LFJ (21.32 [18.27, 26.05]) was higher than those of grade I (18.33 [15.47, 22.16], p < 0.001), grade II LFJ (17.99 [15.18, 20.97], p < 0.001), and grade III LFJ (18.29 [15.07, 25.47], p = 0.178). In four-area method analysis, the median T2* value of grade 0 LFJ (21.55 [18.2, 26.72]) was significantly higher than those of grade I (17.94 [15.45, 21.67], p < 0.001), grade II LFJ (17.28 [14.65, 20.38], p < 0.001) and grade III LFJ (18.25 [15.22, 22.41], p = 0.028). A downward trend in T2* value was observed as LFJ degeneration progressed, except for grade III. Additionally, the median T2* values obtained using all-inclusive method were generally higher than those from four-area method, except for grade 0. Four-area method demonstrated excellent inter-observer reliability with ICC of 0.992 ([0.99, 0.993], p < 0.001), higher than that of all-inclusive method (0.942 [0.931, 0.951], p < 0.001). CONCLUSIONS: Compared to all-inclusive method, four-area method provides higher reproducibility and accuracy in measuring T2* values. Thus, it is a more reliable approach for assessing biochemical changes in LFJ degeneration on T2* mapping.