No abstract

Of 352 patients with colorectal carcinoma followed for a minimum of 5 years after surgery, 39 cases (11.1%; median age 60 years) had mucinous adenocarcinoma, and 4 (1.1%; median age 62 years) had signet-ring cell carcinoma. Mucinous carcinomas were most frequently located in the rectum (61.5%) and in the sigmoid colon (15.3%) and presented with stage C and D disease in 41 and 15% of the cases, respectively. Disease recurrence was more frequently observed in patients with mucinous (51.7%) or signet-ring lesions (100%) as compared with adenocarcinomas. Five-year survival was 45 (median 48 months), 28 (median 27), and 0% (median 15 months) in patients with adenocarcinomas, mucinous adenocarcinomas, and signet-ring cell carcinomas, respectively (p < 0.05). Mucinous carcinomas of the rectum had had a significantly worse prognosis (5-year survival 17%, median 33 months) as compared with adenocarcinomas of the same site (5-year survival 34%, median 25 months; p < 0.05).

The histories of 90 consecutive patients who underwent curative surgery for rectal cancer from January 1975 to December 1981 were reviewed. Twenty of 90 patients (22.2%) recurred locally. The site of the primary lesion and tumor differentiation are the most important factors influencing local recurrence. Median survival from recurrence was 3 months and no patient was alive after 15 months. No significant difference in survival was observed in patients surgically treated for recurrence compared to untreated patients.

In our study, adherence to a strict follow-up program unfortunately proved to be ineffective for improving long-term survival for patients who underwent reoperation with curative intent.

Variables correlated to survival were studied in 121 patients who had undergone potentially curative surgery for left colon and rectal cancer. To investigate the prognostic value of the parameters both univariate and multivariate analysis were carried out. Minimum follow-up was 5 years. Multivariate analysis showed that while disease stage (p < 0.0001) and site of primary tumor (p < 0.0006) independently influenced survival, type of surgical procedure and histopathologic grade had no impact on survival. Jass histopathologic classification predicted survival for patient group I and IV whereas no significant relationship was observed for group II and III, which, in our series, were the most frequently encountered groups. Of the three parameters considered for Jass classification, tubule configuration, pattern of tumor growth and lymphocytic infiltration, only the latter was significantly correlated to survival (p < 0.005). Different results were obtained when the prognostic values of Jass group was investigated separately for the 49 patients with adenocarcinoma of the left colon and the 72 patients with rectal cancer. Further investigation is required before routine clinical application of the Jass classification can be recommended.