Intraoperative ventilation risk factors for postoperative pulmonary complications in patients undergoing elective neoplastic esophagectomy: a retrospective cohort study

• Postoperative pulmonary complications are a major complication after elective neoplasic esophagectomy. • Plateau pressure, respiratory rate, and driving pressure were independently associated with postoperative pulmonary complications during both one- and two-lung ventilation phases • During two-lung ventilation, longer duration increases the risk of postoperative pulmonary complications, whereas higher PEEP is protective of PPCs. • During one-lung ventilation, higher tidal volume and mechanical power are risk factors for postoperative pulmonary complications. • Intraoperative ventilatory parameters may be modifiable targets to reduce pulmonary morbidity after elective neoplastic esophagectomy. Postoperative pulmonary complications (PPCs) are frequent and serious complications after esophagectomy. Our aim was to determine intraoperative ventilatory parameters that are independently associated with PPCs during one-lung ventilation (OLV) and two-lung ventilation (TLV) phases. This retrospective single-center cohort study included 454 patients undergoing elective esophagectomy with combined abdominal and thoracic approaches. The primary outcome was the occurrence of PPCs within seven postoperative days. Separate predictive models were established for OLV and TLV using multivariable logistic regressions with ventilatory parameters. PPCs occurred in 194 patients (42.7%). Plateau pressure (Pplat) and driving pressure (DP) were independently associated with PPCs during both phases (Pplat: OR = 1.73, 95%CI: 1.25-2.38, p < 0.001 in OLV; OR = 1.38, 95%CI: 1.00-1.92, p = 0.05 in TLV; DP: OR = 1.17, 95%CI: 1.05-1.30, p = 0.005 in OLV; OR = 1.14, 95%CI: 1.06-1.23, p < 0.001 in TLV). Respiratory rate (RR) was also associated (OR up to 1.62). In TLV, duration of ventilation (OR between 1.39 and 1.41) was associated with PPCs, while positive end-expiratory pressure (PEEP) was protective (OR 0.73). In OLV only, higher tidal volume (VTe) (OR 1.31, 95%CI: 1.303, 1.66, p = 0.02) and mechanical power (MP) (OR = 1.29, 95%CI: 1.00-1.64, p = 0.05) were associated with PPCs. Model discrimination was acceptable (AUC 0.70–0.76). Pplat, RR, and DP were associated with PPCs in both ventilation phases. During TLV, duration increased and PEEP decreased PPC risk, whereas during OLV, VTe and MP were associated with PPCs.