BACKGROUND: Ten-year survival rates are only rarely reported and frequently include a large proportion of censored data-that is, most of the patients have not survived the 10 years. We therefore selected patients in a prospectively maintained, hospital-based tumor register who had been operated on for colorectal carcinoma (CRC) more than 10 years earlier and who were classified as long-term survivors. METHODS: For 589 consecutive CRC patients who underwent R0 resection in the period 1990-1998, we compared prognosis-relevant characteristics and calculated the survival rate as a function of age, sex, location of the tumor, general state of health, urgency of the operation, and pT and pN class. All patients were observed until their death or until at least 10 years after resection. Patients who died of other causes were censored. Overall survival and relative survival (the latter based on tumor-related death) were assessed. RESULTS: The 10-year survivors were more often female (not significant), younger (p < 0.001), in good general health (p < 0.001), had undergone elective resection (p < 0.001), and had early-stage tumors (p < 0.001). In the univariate analysis emergency operation, impaired general health, invasion beyond the muscularis propria, and lymph-node metastasis were found to reduce relative survival. In the multivariate analysis, location, emergency resection, pT, and pN were found to be statistically independent risk factors. CONCLUSIONS: Long-term freedom from tumor recurrence, like-short-term, is influenced largely by factors that are beneficially influenced by early recognition. The patient's age at resection is immaterial.
BACKGROUND: Ten-year survival rates are only rarely reported and frequently include a large proportion of censored data-that is, most of the patients have not survived the 10 years. We therefore selected patients in a prospectively maintained, hospital-based tumor register who had been operated on for colorectal carcinoma (CRC) more than 10 years earlier and who were classified as long-term survivors. METHODS: For 589 consecutive CRCpatients who underwent R0 resection in the period 1990-1998, we compared prognosis-relevant characteristics and calculated the survival rate as a function of age, sex, location of the tumor, general state of health, urgency of the operation, and pT and pN class. All patients were observed until their death or until at least 10 years after resection. Patients who died of other causes were censored. Overall survival and relative survival (the latter based on tumor-related death) were assessed. RESULTS: The 10-year survivors were more often female (not significant), younger (p < 0.001), in good general health (p < 0.001), had undergone elective resection (p < 0.001), and had early-stage tumors (p < 0.001). In the univariate analysis emergency operation, impaired general health, invasion beyond the muscularis propria, and lymph-node metastasis were found to reduce relative survival. In the multivariate analysis, location, emergency resection, pT, and pN were found to be statistically independent risk factors. CONCLUSIONS: Long-term freedom from tumor recurrence, like-short-term, is influenced largely by factors that are beneficially influenced by early recognition. The patient's age at resection is immaterial.
Authors: Jose G Guillem; David B Chessin; Alfred M Cohen; Jinru Shia; Madhu Mazumdar; Warren Enker; Philip B Paty; Martin R Weiser; David Klimstra; Leonard Saltz; Bruce D Minsky; W Douglas Wong Journal: Ann Surg Date: 2005-05 Impact factor: 12.969