Predicting primary postoperative pulmonary complications in patients undergoing minimally invasive surgery for colorectal cancer.

OBJECTIVE: To determine the risk factors for developing primary postoperative pulmonary complications (PPC) in patients undergoing minimally invasive colorectal surgery (MIS) for the treatment of cancer and to identify the potential indicators for more extensive preoperative evaluation.
MATERIALS AND METHODS: The ACS-NSQIP® database was interrogated to capture patients who had elective colon or rectal cancer and underwent MIS between 2012 and 2017. Patients who had primary PPC including pneumonia, unplanned intubation and/or failure to wean from mechanical ventilation for > 48 h were compared to patients without PPC. Significant risk factors for PPC were retained to build a predictive risk model through logistic regression analysis. The model was then internally validated using 2018 data.
RESULTS: Of 50,150 patients identified, 637 (1.3%) had PPC. The final risk prediction model included six variables: history of chronic obstructive pulmonary disease, age, smoking status, functional health status, pre-operative congestive heart failure, and American Society of Anesthesiology class ≥ 3. The model achieved good calibration (Hosmer-Lemeshow goodness-of-fit test, p = 0.614) and discrimination (c statistics = 0.757). Internal validation achieved similar discrimination (c statistics = 0.798).
CONCLUSION: Primary postoperative pulmonary complications affected 1.3% of patients undergoing MIS for colon or rectal cancer. The novel predictive risk score showed good discrimination and may help to identify patients who may benefit from perioperative optimization.