Search

OBJECTIVE. The purpose of this study was to assess whether MR volumetric data on DW and T2-weighted MR images are correlated with lymphovascular invasion and lymph node metastases in resectable rectal cancer. MATERIALS AND METHODS. This retrospective study consisted of 50 consecutive patients with rectal cancer who underwent radical surgery within 1 week of MRI. The gross tumor volume was determined on both diffusion-weighted and T2-weighted MR images and correlated with pathologic lymphovascular invasion and lymph node metastases using univariate, multivariate, and ROC curve analyses. RESULTS. Both gross tumor volume values showed correlations with lymphovascular invasion ( r = 0.750 vs r = 0.710; p < 0.0001) and lymph node metastases ( r = 0.780 vs r = 0.755; p < 0.0001). Both values were associated with lymphovascular invasion and lymph node metastases in univariate analysis (all p < 0.0001), whereas only the DWI-based value was an independent risk factor for lymphovascular invasion (odds ratio = 1.207; p = 0.005) and lymph node metastases (odds ratio = 1.420; p = 0.005) in multivariate analysis. Both values could distinguish between N0 and N1, N0 and N1-N2, and N0-N1 and N2 disease (all p < 0.0001) in the Mann-Whitney U test. The area under the ROC curve was higher for the DWI-based value in lymphovascular invasion (0.899 vs 0.877), N0 vs N1 (0.865 vs 0.827), N0 vs N1-N2 (0.934 vs 0.911), and N0-N1 vs N2 (0.932 vs 0.927). CONCLUSION. Tumor volumetry data correlated with both lymphovascular invasion and lymph node metastases in resectable rectal cancer. In particular, the DWI-based gross tumor volume showed the most potential for noninvasive preoperative evaluation of lymphovascular invasion and lymph node metastases.