Goh H S, Goh C R, Rauff A, Foong W C
University Department of Surgery, National University of Singapore, National University Hospital.
Ann Acad Med Singap. 1987 Jul;16(3):437-40.
Clinical and pathological data from 219 cases of large bowel cancer were analysed using the Cox regression model to evaluate their effects on survival. Clinical variables such as sex, change in bowel habit, bleeding per rectum, tumour site, size and duration of symptoms had no significant influence on survival. Individually, age, abdominal distension, Dukes' stage and tumour grade had significant influence on survival. When considered together, Dukes' stage was the only variable selected. Therefore Dukes' stage is the most important prognostic factor in colorectal cancer. The distribution of patients was 25, 18, 26 and 31 percent for Dukes' A, B, C and D respectively. The probabilities of survival at five years were 71, 55, 36 and 16 percent for the respective Duke's stages. These data show that the clinical and pathological variables of patients with colorectal cancer in Singapore and the factors significantly associated with survival are similar to those in western countries.