Ventilation Mode and Atelectasis After Prone-Position Surgery: A Lung Ultrasound Study

Overview

This prospective single-centre observational study compared two intraoperative mechanical ventilation modes, volume-controlled ventilation (VCV) and VCV with AutoFlow/AF, in adults undergoing elective surgery in the prone position. The ventilation mode was selected by the attending anaesthesiologist as part of routine clinical care; the investigators did not assign it. Perioperative atelectasis was assessed using a modified bilateral 12-zone lung ultrasound score obtained before induction of anaesthesia and after extubation, and oxygenation was assessed using the alveolar-arterial oxygen gradient calculated from an intraoperative arterial blood gas sample. The study also examined whether the effect of ventilation mode on atelectasis differed according to obesity.

Ninety adults (American Society of Anesthesiologists physical status I-II, body mass index below 35 kg/m2, aged 18 to 65 years) scheduled for elective prone-position surgery were grouped by the ventilation strategy chosen by the attending anaesthesiologist: VCV (n=45) or VCV-AF (n=45). Both groups received a tidal volume of 6 to 8 mL/kg and median positive end-expiratory pressure of 5 cmH2O, delivered with a Draeger Perseus A500 anaesthesia workstation. A modified lung ultrasound score was obtained in the supine position using a 2 to 5 MHz convex probe before induction and at least 30 minutes after extubation, with each hemithorax divided into anterior, lateral and posterior regions and each region scored for loss of aeration. The alveolar-arterial oxygen gradient was calculated from an arterial blood gas sample taken 15 minutes after prone positioning. Patients were stratified by body mass index (below 30 versus 30 kg/m2 or above). The primary comparison was the postoperative lung ultrasound score between the two ventilation modes. Pre-specified analyses included adjustment for the baseline lung ultrasound score using analysis of covariance and a formal ventilation-mode by obesity interaction test. The correlation between the lung ultrasound score and the alveolar-arterial oxygen gradient was also assessed. Of 97 patients initially enrolled, seven were excluded because postoperative lung ultrasound data could not be completed owing to unavailability of the ultrasound device or operator, leaving 90 patients with complete data for analysis.