BACKGROUND Postoperative delirium (POD) is a frequent complication in elderly patients, usually occurring within a few days after surgery. This study investigated the effect of lung-protective ventilation (LPV) on POD in elderly patients undergoing spinal surgery and the mechanism by which LPV suppresses POD. MATERIAL AND METHODS Seventy-one patients aged ≥65 years were randomized to receive LPV or conventional mechanical ventilation (MV), consisting of intermittent positive pressure ventilation following induction of anesthesia. The tidal volume in patients who received MV was 8 ml/kg predicted body weight (PBW), and the ventilation frequency was 12 times/min. The tidal volume in patients who received LPV was 6 ml/kg PBW, the positive end-expiratory pressure was 5 cmH₂O, and the ventilation frequency was 15 times/min, with a lung recruitment maneuver performed every 30 min. Blood samples were collected immediately before anesthesia induction (T₀), 10 min (T₁) and 60 min (T₂) after turning over, immediately after the operation (T₃), and 15 min after extubation (T₄) for blood gas analysis. Simultaneous cerebral oxygen saturation (rSO₂) and cerebral desaturation were recorded. Preoperative and postoperative serum concentrations of interleukin (IL)-6, IL-10 and glial fibrillary acidic protein (GFAP) were measured by ELISA. POD was assessed by nursing delirium screening score. RESULTS Compared with the MV group, pH was lower and PaCO₂ higher in the LPV group at T₂. In addition PaO₂, SaO₂, and PaO₂/FiO₂ were higher at T₁, and T₄, and rSO₂ was higher at T₃, and T₄ in the LPV than in the MV group (P<0.05 each). Postoperative serum GFAP and IL-6 were lower and IL-10 higher in the LPV group. The incidences of cerebral desaturation and POD were significantly lower in the LPV group (P<0.05). CONCLUSIONS LPV may reduce POD in elderly patients undergoing spinal surgery by inhibiting inflammation and improving cerebral oxygen metabolism.