Safety and feasibility of three-dimensional McKeown minimally invasive esophagectomy.

BACKGROUND: To compare the perioperative outcomes from McKeown minimally invasive esophagectomy (MIE) when performed in three-dimensional versus two-dimensional visualization system, and investigate the learning curve of a single surgeon who implemented three-dimensional McKeown MIE. METHODS: A total of 335 consecutive cases (three-dimensional or two-dimensional) were identified. Perioperative clinical parameters were compared and cumulative sum learning curve was plotted. Propensity score matching was used to reduce selection bias from confounding factors. RESULTS: Patients in three-dimensional group were associated with more chronic obstructive pulmonary disease (23.9% vs 3.0%, p < 0.01). After propensity score matching (108 matched patients in each groups), this finding was no longer statistically significant. Comparing to two-dimensional group, significant improvement in total retrieved lymph nodes (28 vs 33, p = 0.003) was observed in three-dimensional group. In addition, more lymph nodes around the right recurrent laryngeal nerve were harvested in three-dimensional group than that in two-dimensional group (p = 0.045). However, there were no significantly differences were found between the two groups in terms of other intraoperative parameters (e.g., operative time) and postoperative relevant outcomes (e.g., lung infection). Furthermore, the change point in the cumulative sum learning curves for intraoperative blood loss and thoracic procedure time was 33 procedures, respectively. CONCLUSION: Three-dimensional visualization system appears to be superior in performing lymphadenectomy during McKeown MIE to that of a two-dimensional technique. For surgeons proficient in performing two-dimensional McKeown MIE, the learning curve for a three-dimensional procedure appears to begin near proficiency after more than 33 cases.