Pre- and intra-operative risk factors predict postoperative respiratory failure after minimally invasive oesophagectomy.

OBJECTIVES:Severe pulmonary complications such as postoperative respiratory failure can occur after minimally invasive oesophagectomy. However, the risk factors have not been well identified. The goal of this study was to develop a predictive model for the occurrence of postoperative respiratory failure with a large sample. METHODS:We collected data from patients with oesophageal cancer who had a minimally invasive oesophagectomy at Shanghai Chest Hospital from 2019 to 2022. Univariable and backward stepwise logistic regression analysis of 19 pre- and intra-operative factors was used before model fitting, and its performance was evaluated with the receiver operating characteristic curve. Internal validation was assessed with a calibration plot, decision curve analysis and area under the curve with 95% confidence intervals, obtained from 1000 resamples set by the bootstrap method. RESULTS:This study enrolled 2,386 patients, 57 (2.4%) of whom developed postoperative respiratory failure. Backward stepwise logistic regression analysis revealed that age, body mass index, cardiovascular disease, diabetes, diffusion capacity of the lungs for carbon monoxide, tumour location and duration of chest surgery ≥101.5 min were predictive factors. A predictive model was constructed and showed acceptable performance (area under the curve: 0.755). The internal validation with the bootstrap method proves the good agreement for prediction and reality. CONCLUSIONS:Obesity, severe diffusion dysfunction and upper segment oesophageal cancer were strong predictive factors. The established predictive model has acceptable predictive validity for postoperative respiratory failure after minimally invasive oesophagectomy, which may improve the identification of high-risk patients and enable health-care professionals to perform risk assessment for postoperative respiratory failure at the initial consultation.