Trends of Minimally Invasive Surgery and Adequate Lymph Node Harvest in pT4 Colon Adenocarcinoma

Abstract

Background: Recent advances in minimally invasive surgery (MIS) have improved outcomes in colon cancer treatment. However, the use of MIS in pT4 colon cancer is controversial. Existing literature on lymph node retrieval in MIS compared to that in open surgery is inconsistent. Previous studies using the National Cancer Database from 2010 to 2014 and 2010 to 2016 found that MIS was associated with higher odds of adequate lymph node resection. Objectives: Using data from more recent years, this study seeks to compare lymph node resection between MIS and open surgery in pT4 colon cancer. Design and Setting: This retrospective cohort study was conducted using the National Cancer Database. Patients and Methods: Patients undergoing MIS or open surgery with lymph node harvest for pT4 colon cancer were included in this study. A trend analysis of surgical approach (MIS versus open and robotic versus laparoscopic) from 2010 to 2017 and a logistic regression model including multiple tumor characteristics and demographic variables for procedures between 2015 and 2017 were performed. Main Outcome Measures: The main outcome evaluated was the adequacy of lymph node retrieval. Sample Size: A total of 27319 patients were included. Results: This study showed a trend towards MIS and robotic surgery for pT4 colon cancer. In addition, it found that MIS for pT4 tumors have higher odds of achieving an adequate lymph node harvest than open surgery in recent years (2015–2017). The demographic factors associated with inadequate lymph node harvest include race and location of the facility (rural versus metro cities). Conclusions: The findings of this study suggest that colorectal surgery has shifted towards MIS, and specifically, robotic surgery for pT4 tumors. This approach has an oncological advantage of adequate lymph node harvest compared to open surgery. Limitations: Despite its advantages, this retrospective database study has inherent biases, such as confounding bias, selection bias, and coding errors in the database, as well as limited data available for analysis. Conflict of interest: The authors have no conflict of interest to declare.