Comparison of restrictive fluid therapy with goal-directed fluid therapy for postoperative delirium in patients undergoing spine surgery: a randomized controlled trial.

Background: Postoperative delirium (POD) is a common phenomenon after spinal surgery. Intraoperative fluid management may affect POD. The aim of this study was to compare the effects of restrictive fluid therapy (RF) with those of goal-directed fluid therapy (GDT) on POD.

Methods: A total of 195 patients aged ≥ 50 years who underwent spinal surgery were randomly divided into two groups: the RF group and the GDT group. In group RF, a bolus of lactated Ringer's solution was administered at a dose of 5 mL·kg before the induction of anesthesia, followed by a dose of 5 mL·kg·h until the end of surgery. For patients in the GDT group, in addition to the initial administration of lactated Ringer's solution at 5 mL·kg, the subsequent fluid therapy was adjusted by using a continuous noninvasive arterial pressure (CNAP) monitoring system to maintain pulse pressure variation (PPV) ≤ 14%. The primary endpoint was the incidence of POD, assessed once daily with the Confusion Assessment Method-Chinese Reversion (CAM-CR) scale at 1-3 days postoperatively. The secondary endpoints were intraoperative fluid infusion volume, urine volume, mean arterial pressure (MAP), heart rate (HR), cardiac index (CI), regional cerebral oxygen saturation (rSO) value, lactic acid value, and visual analog scale (VAS) pain score at 1-3 days after surgery. Moreover, postoperative complications and the length of hospital stay were recorded.

Results: The incidence of POD was lower in the GDT group than in the RF group (12.4% vs 4.1%; P = 0.035) in the first 3 days after spine surgery. Compared to group RF, group GDT exhibited a significantly increased volume of intraoperative lactated Ringer's solution [1500 (interquartile range: 1128 to 1775) mL vs 1000 (interquartile range: 765 to 1300) mL, P < 0.001] and urine volume [398 (interquartile range: 288 to 600) mL vs 300 (interquartile range: 200 to 530) mL, P = 0.012]. Intraoperative MAP, CI and rSO values were higher in the GDT group than in the RF group (P < 0.05). Moreover, the length of hospital stay [17.0 (14 to 20) days versus 14.5 (13 to 17.0) days, P = 0.001] was shorter in the GDT group than in the RF group.

Conclusions: GDT reduced the incidence of POD in middle- and old-aged patients undergoing spinal surgery possibly by stabilizing perioperative hemodynamic and improving the supply and demand of oxygen.

Trial Registration: ChiCTR2000032603 ; Registered on May 3, 2020.