Obstruction and perforation in colorectal adenocarcinoma: an analysis of prognosis and current trends.

BACKGROUND

In adenocarcinoma of the colon and rectum, obstruction and perforation may occur either alone or together at the site of the neoplasm or proximally. Both events carry a poor prognosis. This retrospective study sought to determine whether a correlation exists between perforation and obstruction, and between these conditions and different clinicopathological factors in colorectal adenocarcinoma.

METHODS

The medical records of 1950 patients with colorectal adenocarcinoma treated in our hospital during a 7-year period were retrospectively analyzed. One hundred patients (5%) were excluded from this study because of a loss of follow-up. Data on clinicopathological factors including age, sex, tumor location, surgical mortality, pathological type, stage, and long-time cancer-free rate were simultaneously analyzed. Patients were grouped as follows: Group 1, complete colonic obstruction without perforation (n = 120). Group 2, complete obstruction with perforation at the site of the cancer (n = 35); Group 3, complete obstruction with perforation proximal to the cancer (n = 13); and Group 4, nonobstructing, nonperforated cancers (n = 1682).

RESULTS

When compared with Group 4, Group 1 had a more advanced Dukes' stage, older age, greater incidence of colonic versus rectal cancers, and a poorer cancer-free survival (P < or = .005). Groups 2 and 3 had a greater incidence of colonic versus rectal cancers (P < or = .004), and Group 3 had a greater operative mortality (P < .001). No significant differences were found between Groups 1, 2, and 3. Multivariate analysis revealed that the independent factors favorable to cancer-free survival (> 5-year survival) were female gender (P = .035), well-differentiated pathology (P < .001), uncomplicated cases (P = .004), colon versus rectal location (P < .001), and early stage (P < .001).

CONCLUSIONS

The perioperative mortality rate for perforated colorectal cancer at the site of the cancer was 9%; for obstructive colorectal cancer, 5%. Perioperative mortality was much greater for perforations of the colon and rectum occurring proximal to the cancer (31%). Survival was worse (P < .001) for patients with obstruction (33%) or perforation proximal to the cancer (33%). The site of perforation did not appear to impact the 5-year survival, although the numbers are relatively small.