Minimally invasive oesophagectomy in the prone versus lateral decubitus position: a systematic review and meta-analysis

The thoracic phase of minimally invasive esophagectomy was initially performed in the lateral decubitus position (LDP); however, many experts have gradually transitioned to a prone position (PP) approach. The aim of the present systematic review and meta-analysis is to quantitatively compare the two approaches. A systematic literature search of the MEDLINE, Embase, Google Scholar, Web of Knowledge, China National Knowledge Infrastructure and databases was undertaken for studies comparing outcomes between patients undergoing minimally invasive esophageal surgery in the PP versus the LDP. In total, 15 studies with 1454 patients (PP; n = 710 vs. LDP; n = 744) were included. Minimally invasive esophagectomy in the PP provides statistically significant reduction in postoperative respiratory complications (Risk ratios 0.5, 95% confidence intervals [CI] 0.34-0.76, P < 0.001), blood loss (weighted mean differences [WMD] -108.97, 95% CI -166.35 to -51.59 mL, P < 0.001), ICU stay (WMD -0.96, 95% CI -1.7 to -0.21 days, P = 0.01) and total hospital stay (WMD -2.96, 95% CI -5.14 to -0.78 days, P = 0.008). In addition, prone positioning increases the overall yield of chest lymph node dissection (WMD 2.94, 95% CI 1.54-4.34 lymph nodes, P < 0.001). No statistically significant difference in regards to anastomotic leak rate, mortality and 5-year overall survival was encountered. Subgroup analysis revealed that the protective effect of prone positioning against pulmonary complications was more pronounced for patients undergoing single-lumen tracheal intubation. A head to head comparison of minimally invasive esophagectomy in the prone versus the LDP reveals superiority of the former method, with emphasis on the reduction of postoperative respiratory complications and reduced length of hospitalization. Long-term oncologic outcomes appear equivalent, although validation through prospective studies and randomized controlled trials is still necessary.