Magnetic resonance staging in rectal cancer patients before total mesorectal excision: Efficacy of MR imaging with body coil, small FOV and thin section parameters
Neşenur Yaran1, Arzu Arslan1, Ahmet Alponat2, Gür Akansel1, Yeşim Gürbüz3, Nagihan İnan1, Tahsin Sarısoy1, Yonca Akgöz1, Ali Demirci1
1Kocaeli Üniversitesi Tıp Fakültesi Radyoloji AD, KOCAELİ
2Kocaeli Üniversitesi Tıp Fakültesi Genel Cerrahi AD, KOCAELİ
3Kocaeli Üniversitesi Tıp Fakültesi Patoloji AD, KOCAELİ
Purpose: To evaluate the efficacy of thin slice and small FOV MRI technique using body coil, performed after distending the rectum with saline solution in the preoperative staging of rectum cancer.
Materials and Methods: MR imaging of 36 patients with rectum cancer was done by distending the rectum with saline solution, using body coil, with thin slices and small FOV parameters. MR staging was done according to the TNM system by consensus of two radiologists using the MR findings of local tumoral invasion, distance to the mesorectal fascia and involvement of local and distant lymph nodes. 12 patients who weren't operated were excluded from the study. 24 patients (13 men, 11 women) who underwent operation with total mesorectal excision technique were included in the study. Histopathologic staging of 24 patients were T1,2 and 3 in 2, 10 and 12; and N0,1 and 2 in 13, 7 and 4 patients, respectively. Comparing MR staging results with the histologic staging, sensitivity, specificity, accuracy and predictive values of the MR technique were determined.
Results: The accuracy of the MRI for determining the T and N stages was 70.8% (17/24) and 58.3% (14/24), respectively. In T staging, accuracy of MR for stages T1, 2 and 3 were 100% (2/2), 70% (7/10) and 66.7 % (8/12), respectively. For stages N0, 1 and 2, accuracy of MR was 53.7% (7/13), 85.7% (6/7) and 25% (1/4), respectively. In determining the T stage, overstaging and understaging rates were 12.5% (3/24) and 16.7% (4/24), respectively. For lymph node involvement sensitivity, specificity, positive predictive value (PPV) and the negative predictive value (NPV) were 90.9% (10/11), 53.8% (7/13), 62.5% (10/16) and 87.5% (7/8), respectively. For determination of tumor deposits accuracy, sensitivity, specificity and NPV were 83.3% (20/24), 20% (1/5), 100% (19/19) and 82.6% (19/23), respectively. Accuracy for the distance to the mesorectal fascia in T3 tumors was 75% (6/8). Accuracy of circumferential resection margin was 66.7% (4/6) and 100% (2/2) in patients with a positive and negative margin, respectively.
Conclusion: The small FOV and thin section MR technique using body coil was efficient in staging of rectal cancers. To provide rectal distension with saline was useful for distinguishing stage T2 and T3 tumors.
Keywords: Rectal cancer, preoperative staging, MR