Population-based esophageal cancer survival after resection without neoadjuvant therapy: An update

Background. There are few population-based studies addressing the survival after resection for esophageal cancer. This study represents an update of a nationwide Swedish cohort initiated in 1987. Methods. Based on data from the Swedish Patient Register, Swedish Cancer Register, and histopathologic records, 1,008 patients who had undergone esophageal resection as the only treatment for esophageal cancer were identified between January 1, 1987 and December 31, 2005. These were followed until death or emigration through linkage to the Swedish Total Population Register until January 1, 2009. Tumor stage, location, and histology were assessed from histopathologic reports, and comorbidities were assessed from the Patient Register. Cox proportional hazards regression models were used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs) regarding survival. The results were adjusted for age, sex, comorbidity, tumor stage, location, histology, surgical radicality, and hospital volume. Results. The proportion of patients surviving for 5 years increased from 19.7% in 1987-1991 to 30.7% in 1997-2000, but remained at 30.5% between 2001 and 2005. No difference in overall adjusted survival was found between the periods of 2001-2005 and 1997-2000 (adjusted FIR, 0.89; 95% CI, 0.70-1.13). Thirty-day mortality decreased from 4.9% in 1997-2000 to 2.0% in 2001-2005, rendering an adjusted HR of 0.26 (95% CI, 0.08-0.87). Conclusion. After adjusting for relevant prognostic factors, long-term population-based survival after resection for esophageal cancer was unchanged between 2001 and 2005 compared to 1997-2000, while the corresponding 30-day mortality improved. (Surgery 2012;152:903-10.)