Objective

To determine if histograms of ADC can be used to differentiate orbital lymphoma,pleomorphic adenoma(PA) and adenoid cystic carcinoma(ACC).

Materials

45 patients with confirmed histologically orbital lymphoma in 20(44.45% ) patients,PA in 15(33.33% ) patients,and ACC in 10(22.22%) patients were enrolled in the study,accepting MRI at 3T. The ROIs were drawn on the maximum level of ADC,and the histogram analysis were performed using the software named Mazda.One-way ANOVA and Kruskal-Wallis test were performed on the nine histogram parameters to find out the different characteristics.LSD and Bonferroni test were used between any two of the three groups and the ROC curve was drawn to evaluate the diagnostic efficacy.

Results

Histogram metrics derived from mean,Perc.01%,Perc.10%,Perc.50%,Perc.90% and Perc.99% were significantly lower in lymphoma than ACC. Lymphoma was significantly lower mean,Perc.01%,Perc.10%,Perc.50% and Perc.90% than PA. Optimal diagnostic performance to predict lymphoma from ACC ( AUC= 0.96,sensitivity = 100.00%,specificity =86.70%) was obtained when setting Perc.99%= 132.00 as the threshold value. A threshold value of 95.50 for Perc.50% had the best diagnostic efficacy to predict lymphoma from PA(AUC= 0.80,sensitivity = 70.00%,specificity =100.00%).

Conclusions

The ADC histogram analysis may help to discriminate orbital tumors of lymphoma,PA and ACC.

Acknowledgements

No acknowledgement found.

References

No reference found.