Résumé
In spite of recent advances in our knowledge of tumor biology and therapy, the management and prognosis of patients with colon cancer (CC) revealed by intestinal obstruction or peritonitis (IOP) are not well defined. This study was undertaken to identify other factors which may influence morbidity and prognosis of CC revealed by IOP.BACKGROUNDIn spite of recent advances in our knowledge of tumor biology and therapy, the management and prognosis of patients with colon cancer (CC) revealed by intestinal obstruction or peritonitis (IOP) are not well defined. This study was undertaken to identify other factors which may influence morbidity and prognosis of CC revealed by IOP.We studied medical documentation from 247 incidental cases of CC registered in the Herault district in 1992 and followed up until 1997. The difference in postoperative mortality and the 5-year survival rate between CC with (n=41) and without IOP (n=206) were evaluated separately in relation to clinico-pathological characteristics, surgeon volume, and healthcare institutions by non-conditional logistic regression and multivariate survival analysis.MATERIAL/METHODSWe studied medical documentation from 247 incidental cases of CC registered in the Herault district in 1992 and followed up until 1997. The difference in postoperative mortality and the 5-year survival rate between CC with (n=41) and without IOP (n=206) were evaluated separately in relation to clinico-pathological characteristics, surgeon volume, and healthcare institutions by non-conditional logistic regression and multivariate survival analysis.Higher rates of Dukes' stages B-C-D were observed in the emergency group than in the elective. Older age was associated with the presence of IOP (p=0.02). Factors predictive for postoperative mortality were IOP (p=0.02), surgical treatment in non-specialized hospitals (p=0.04) and residence in rural areas (p=0.01). The primary independent negative prognostic factor in multivariate analysis was the presence of nodal or distant metastases (Dukes' stage D, p=0.0001), followed by lack of chemotherapy (p=0.008), initial treatment in non-specialized hospitals (p=0.01), onset with IOP (p=0.02), low-volume surgeons (p=0.02).RESULTSHigher rates of Dukes' stages B-C-D were observed in the emergency group than in the elective. Older age was associated with the presence of IOP (p=0.02). Factors predictive for postoperative mortality were IOP (p=0.02), surgical treatment in non-specialized hospitals (p=0.04) and residence in rural areas (p=0.01). The primary independent negative prognostic factor in multivariate analysis was the presence of nodal or distant metastases (Dukes' stage D, p=0.0001), followed by lack of chemotherapy (p=0.008), initial treatment in non-specialized hospitals (p=0.01), onset with IOP (p=0.02), low-volume surgeons (p=0.02).As the presence of IOP influenced independently postoperative mortality and specific five-year survival, we suggest that the management of CC revealed by IOP should be concentrated in the hands of those surgeons and healthcare institutions that show the best outcomes.CONCLUSIONSAs the presence of IOP influenced independently postoperative mortality and specific five-year survival, we suggest that the management of CC revealed by IOP should be concentrated in the hands of those surgeons and healthcare institutions that show the best outcomes.