PWE-147 The impact of body mass index on outcomes following oesophagectomy in western patients: a meta-analysis

Abstract

Introduction Multiple groups have previously reported the influence of BMI on outcomes following oesopheageal surgery but their conclusions have been limited by a lack of clear delineation of patient characteristics as well as geographical location and the nature of multimodal treatment. In order to address the specific question concerning the impact of preoperative BMI on western patients’ outcomes after oesophageal cancer surgery we have performed a meta-analysis, taking into account histology characteristics wherever possible. Method Between December 2014 and January 2015 a comprehensive electronic search was performed to identify Western publications relating to BMI and outcomes after oesophagectomy. Articles that did not report preoperative BMI, morbidity, early mortality and 5 years overall survival were excluded. The search was performed using Pubmed and the reference lists for each of the articles included was also reviewed in order to find any additional studies. Statistical analyses were performed using OpenMetaAnalyst software (Version 10.10). Results 161 records were examined, 9 western studies were included with a total of 3150 pts of which 317 were specifically affected by adenocarcinoma. The study population was stratified into 2 groups: normal group (BMI 30). Whilst higher BMI was associated with a greater risk for anastomotic leaks (near 25% with a p-value of 0.102), bleeding (18% and 0.213), wound infections (25% and 0.090) and poorer long term survival (near 20% and 0.090) this was not found to be statistically significant. Cardiac, pulmonary and any major complications were not significantly different between the 2 groups. On the other hand, there was a 20% greater risk for early postoperative mortality in the normal BMI group although the difference did not reach statistical significance as well as recurrent laryngeal nerve injury. Normal BMI was also significantly associated with chyle leaks (0.021). Stratification according to histology was possible in the analysis of 3 postoperative complications (anastomotic leakages, pulmonary events and early mortality). In the adenocarcinoma group High BMI was significantly associated with pulmonary complications (0.009). Conclusion This meta-analysis shows that in Western studies BMI above 30 is not associated with increased mortality or morbidity after oesophageal surgery, although there is an increase in pulmonary complications of those who have adenocarcinoma. BMI These results suggest that BMI should not exclude patients from being treated with curative intent in multimodal treatment; however the increased risk of pulmonary complications should trigger careful preoperative respiratory assessment and optimisation. Disclosure of interest None Declared.