Abstract
OBJECTIVE: Multidisciplinary treatment is the standard of care for patients with cancer of the rectum. Neoadjuvant therapy decisions rely on the local carcinoma stage's magnetic resonance imaging assessment. However, the reliability of magnetic resonance imaging in evaluating mesorectal lymphadenopathy remains highly questionable. METHODS: A prospective clinical study was conducted to assess the accuracy of preoperative magnetic resonance imaging in assessing T-stage and N-stage rectal carcinoma. The magnetic resonance imaging findings were compared with those of a detailed histopathological examination of rectal specimens in each patient. RESULTS: The study included 48 patients with cancer of the rectum who had undergone rectal resection with total mesorectal excision. The magnetic resonance imaging assessment of the T-stage was accurate in 26 out of the 48 cases (54.2%). Overstaging of the T-stage was recognized in 18 patients (37.5%), and 4 (8.3%) were understaged. According to preoperative magnetic resonance imaging (using Horvat's criteria), 56 suspicious/malignant lymph nodes were identified. However, only 13 (23.2%) of these lymph nodes were proved malignant on histopathology examination. The sensitivity of magnetic resonance imaging for nodal involvement assessment was 52.0% and specificity was 93.6%. Of the 646 magnetic resonance imaging-negative lymph nodes, 12 (1.9%) were false negatives (histopathology revealed metastatic lymph nodes). Of note, 20 study patients with malignant lymphadenopathy (according to magnetic resonance imaging) were detected. The magnetic resonance imaging diagnosis of malignant lymphadenopathy was correct in only 30.0% of these patients. CONCLUSION: The accuracy of magnetic resonance imaging in evaluating mesorectal lymph nodes for cases with cancer of the rectum is remarkably low. Therefore, neoadjuvant therapy indications should be based primarily on the magnetic resonance imaging description of the depth of carcinoma invasion.