Postoperative Outcomes Following Prolonged One-Lung Ventilation in Thoracic Surgery in Relation to Intraoperative...

To evaluate the association between intraoperative ventilation parameters during prolonged one-lung ventilation (OLV) and postoperative outcomes.

Retrospective multicenter cohort study.

German Thorax Registry.

Two thousand twenty-six patients who underwent thoracic surgery with OLV >60 minutes between 2016 and 2021.

None.

Parameters included peak pressure-derived driving pressure (DP), calculated as maximum airway pressure (pMax) minus positive end-expiratory pressure (PEEP); pMax; PEEP; and tidal volume (TV) per predicted body weight (PBW). Primary outcomes were overall postoperative complications and respiratory complications; in-hospital mortality was analyzed as a secondary outcome. Postoperative complications occurred in 34.1% of patients; respiratory complications, in 18.6%; and in-hospital mortality, in 1.8%. In unadjusted analyses, higher overall postoperative complication rates were observed above several ventilation thresholds, including DP >20 millibars (mbar), pMax >25 and >30 mbar, PEEP >7 and >8 mbar, and TV/PBW >4 and >5 mL/kg PBW (all p < 0.05). Higher overall postoperative complication rates also were observed for OLV duration >180 minutes compared with 61 to 90 minutes and for surgery duration >180 minutes compared with 31 to 60 minutes (both p < 0.001). Respiratory complications were significantly associated with DP >20 mbar (p = 0.009; DP >25 mbar: p = 0.013), pMax >25 mbar (p < 0.001; pMax >30 mbar: p = 0.022), and TV >5 mL/kg PBW (p = 0.032). Respiratory complication rates also were higher for OLV duration >180 minutes compared with 61 to 90 minutes and for surgery duration >180 minutes compared with 31 to 60 minutes (both p < 0.001). In-hospital mortality was significantly associated with DP >20 mbar (p = 0.023), pMax >25 mbar (p = 0.017; pMax >30 mbar: p < 0.001), PEEP >8 mbar (p = 0.027), and TV >8 mL/kg PBW (p = 0.034), but analysis was limited by the low number of events (n = 37).

Several pressure-, volume-, and duration-related thresholds were associated with higher postoperative morbidity and in-hospital mortality in unadjusted analyses. These findings should not be interpreted as evidence to reduce PEEP routinely during OLV. In adjusted models, baseline patient risk and case complexity markers were the main contributors, suggesting that higher PEEP may partly reflect confounding by indication, intraoperative physiology, and procedural complexity. Prospective validation is warranted.