Background: This study was performed to characterize the clinicopathological features of colorectal tumors with flat-, depressed-, or protruded-type morphology (hereafter referred to simply as flat, depressed, or protruded lesions).
Methods: There are two major types of colorectal tumor: polypoid (protruded) and nonpolypoid (flat and depressed). A total of 130 lesions from 130 patients with colorectal submucosal invasive cancer were classified into three groups according to their macromorphology seen during endoscopy: flat (laterally spreading) and depressed nonpolypoid tumors and protruded polypoid tumors. The following factors in the patients' background were evaluated: indication for colonoscopy, age, and family history of colorectal cancer in first-degree relatives (i.e., parents, siblings, children). We also compared the following characteristics of the tumors: size, location, depth of submucosal invasion, vascular invasion, and frequency of synchronous and metachronous tumor lesions.
Results: The incidence of abnormal findings on follow-up studies after polypectomy as an indication for colonoscopy was significantly higher among patients with flat lesions (4/24, 16.7%) and depressed lesions (3/22, 13.6%) than among those with protruded lesions (1/84, 1.2%) (P < 0.01, P < 0.01). Patients with flat lesions (65.8 +/- 7.6 years old) were significantly older than those with protruded lesions (P < 0.05). The patients with flat tumors had a significantly higher rate of a family history of colorectal cancer (6/24, 25.0%) than patients with protruded or depressed lesions (P < 0.01, P < 0.05). The protruded lesions were significantly larger than the depressed lesions (size 13.3 +/- 6.7 mm) (P < 0.05), and the flat lesions (24.1 +/- 10.1 mm) were significantly larger than either the protruded or depressed lesions (P < 0.01, P < 0.01). Seventy-five percent (18/24) of the flat lesions were located in the right colon, and this proportion was significantly higher than that among the protruded or depressed lesions (P < 0.01, P < 0.01). The mean +/- SD depth of submucosal invasion was 1218 +/- 1034 microm in the flat lesions, 2392 +/- 1869 microm in the depressed lesions, and 2761 +/- 1929 microm in the protruded lesions, representing a significant difference (P < 0.05, P < 0.0001). Of the 24 patients with flat lesions, 9 (37.5%) showed vascular invasion; this proportion was significantly lower than that among patients with the depressed or protruded lesions (P < 0.01, P < 0.01). Patients with depressed lesions tended to have higher incidence of synchronous and metachronous malignant polyps than those with protruded or flat lesions.
Conclusion: It is important to examine the morphology of colorectal tumors when diagnosing them and planning the treatment strategy, including follow-up, after resection of nonpolypoid tumors. It is useful to know the patient's family history so nonpolypoid tumors can be accurately diagnosed.