BACKGROUND
CEA is widely used in the follow up of patients with colorectal carcinoma.
AIM
To study the value of preoperative CEA as an independent prognostic factor in colorectal carcinoma.
PATIENTS AND METHODS
Analysis of 373 operated patients (204 females, age range 21-92 years) with colorectal carcinoma and a mean follow up of 53 months. The cutoff value for CEA was 5 ng/ml. Ninety four percent of patients had an excisable tumor, 79% had involvement of perirectal/pericolonic adipose tissue and 46% had lymph node involvement. Staging was done using Dukes-Turnbull and TNM classifications.
RESULTS
CEA was normal in 61% of cases, over 5 ng/ml in 39% and over 15 ng/ml in 22%. There was a strong correlation between mean preoperative CEA and tumor stage, depth and lymph node involvement. During the follow up, 140 patients died, 57 with normal and 83 with elevated CEA. Cancer mortality in patients subjected to a curative excision of the tumor (Dukes A-C2/TNM I-III) was 9% for colonic tumors and 36% for rectal tumors (p < 0.001). There were no survival differences in patients with Dukes B/TNM II tumors according to preoperative CEA. Among Dukes C/TNM III tumors, survival difference was only significant for rectal tumors. A Cox model disclosed tumor stage, location and preoperative CEA as independent prognostic factors for survival.
CONCLUSIONS
CEA is an independent prognostic factor for survival in colorectal carcinoma and high levels suggest an advanced disease.