O19 OUTCOMES OF IVOR LEWIS VERSUS MCKEOWN OESOPHAGECTOMY FOR CANCER: A PROPENSITY SCORE MATCHED ANALYSIS OF THE NETHERLANDS CANCER REGISTRY

Abstract

Abstract Aim The aim this study was to investigate possible differences in outcomes between Ivor-Lewis (IL) versus McKeown esophagectomy (MK) in a population based cohort. Background It is unknown whether IL or MK should be preferred in patients with potentially curable locoregionally oesophageal or gastro-oesophageal junction (GEJ) cancer. Methods Oesophageal and GEJ cancer patients without distant metastases who underwent a transthoracic oesophagectomy between 2015 and 2017 were selected from the Netherlands Cancer Registry. Patient and tumour characteristics, and treatment outcomes were analysed and compared between IL and MK after propensity score matching (PSM). Patients were matched based on sex, age, tumour morphology and topography, comorbidities, neoadjuvant treatment, cT and cN stage, ASA-score and hospital volume. Postoperative complications and survival were compared between both groups using Chi-Square and multivariable Cox regression analyses. In the multivariable Cox regression analyses we corrected for infectious complications (wound complications and pneumonia) and anastomotic leakage to investigate its effect on survival. Results Off all 1764 included patients (1201 IL patients and 563 MK patients), 948 patients remained after matching. Significantly more lymph nodes were yielded after IL compared to MK (P=0.02). Surgical radicality did not differ between both groups (P=0.25). The complication rate was significantly lower in the IL group regarding anastomotic leakage (17% vs. 28%, P<0.0007) and recurrent laryngeal nerve palsy (2% vs. 6%, P<0.001). Cardiac complications were seen more frequently in patients who underwent IL (17% vs. 11%, P<0.02). No differences were observed in 30 day and 90 day mortality. Kaplan-Meier showed a better survival for IL (63%) compared to MK (58%) (log-rank p=0.07). When correcting for anastomotic leakage and infectious complications the difference in survival remained (HR=0.84, 95% CI 0.69 – 1.03). Similar results regarding postoperative complications and survival were observed in the unmatched cohort. Conclusion Patients who underwent an IL have better outcomes in terms of anastomotic leakage and recurrent laryngeal nerve palsy rate and overall survival compared to MK. The difference in survival appears not to be related to the differences in infectious complications. Based on these results, IL appears to be the treatment of choice in patients in whom both oncological resections are appropriate.