Have Multimodal and Perioperative Therapeutic Strategies Changed the Survival after Oncological Oesophageal Surgery in the Last Two Decades? A Retrospective Cohort Analysis over 23 Years

Background Esophageal malignoma is among the most frequent causes for cancer-related deaths. The only definite curative therapy is esophagectomy embedded in various multimodal treatment regimens. The aim of this study was to evaluate long-term development after esophageal surgery in the last two decades in order to observe possible trends and their influence on short, medium and long term survival. Patients and Methods Cases of 301 patients who underwent esophagectomy between 1989 and 2012 were analysed retrospectively. To investigate possible changes in perioperative management and their influence on prognosis patients were divided into two cohorts (A: surgery between 1989 and 2000; B: surgery between 2001 and 2012) and further analyzed with regard to demographics, tumor entity, stage, complications and survival. Statistics were conducted to compare both groups while p0.05 was regarded as statistically significant. Results In cohort B patients were significantly older compared to cohort A and underwent surgery in earlier tumor stages with a higher lymphnode yield. Also an increased incidence of adenocarcinoma was observed. While overall morbidity did not change significantly, a decreased rate of anastomotic leakage was observed in cohort B (5.5%) compared to cohort A (12.3%) accompanied by a simultaneous increase in cardiac events (A: 3.6% vs. B: 12.3%). Overall 30-days-mortality was 2.7% and decreased significantly from 5% in cohort A to 0.7% in cohort B (p=0.05). Median survival was 46 +/- 7 month in cohort A, in cohort B an increase could be observed (53 +/- 7 months, p=0.03). By univariate analysis we could demonstrate that stage, affected lymph nodes, lymphnode ratio (LNR) and incidence of postoperative complications were significant predictors for the survival whereas in multivariate analysis T-stage, R-status and LNR were independent predictors for patients outcome. Conclusion Patients undergoing esophageal resection for cancer nowadays are older than in the past decades. Earlier cancer diagnosis, more radical surgical techniques with an extended lymphnode dissection, a decrease in anastomotic leakage and an improved perioperative care seem to compensate for this potential demographic disadvantage. The most important independent predictor of outcome after esophageal resection is the LNR.