Effect of incremental chest wall constriction on the spatial distribution of voxel-wise lung aeration in an ovine model of ards: a pilot study of the chest-hug project

Position du problème et objectif(s) de l’étude

During open-lung mechanical ventilation, regional lung hyperinflation has been associated with ventilator-induced lung injury. In ARDS patients, clinical settings consistent with limiting hyperinflation through potential effects on chest wall mechanics (e.g., prone position, obesity) have been associated with improved clinical outcomes. This study aims to evaluate the spatial distribution of voxel-wise lung aeration associated with incremental chest wall constriction (CWC) in an ovine model of ARDS when high positive end-expiratory pressure (PEEP) is applied.

Matériel et méthodes

This preclinical study was approved by our institutional ethics committee (#40593_2023011916101935_v2). Ten female sheep (mean weight=20kg) were exposed to experimental ARDS by combining E. Coli endotoxemia and 90 minutes of high-strain ventilation (PEEP=0cmH2O and driving pressure=30cmH2O). PEEP was initially set at 5cmH2O (Baseline5), then increased to 15cmH2O (Baseline15). While maintaining PEEP at 15cmH2O, CWC was performed using a non-extensible belt, with constriction levels monitored using pressure lines connected to saline bags placed between the belt and the animals' chests. Three incremental 30-minute steps of CWC were explored: 30mmHg (P30), 60mmHg (P60), and 90mmHg (P90). Tidal volume was maintained at 8ml/kg throughout the experiment. Multimodal respiratory mechanics were continuously recorded (Fluxmed®). End-expiratory lung CT scans were performed at the end of each step. Following lung segmentation powered by machine learning, local aeration was quantified by computing voxel-wise gas fraction (Fgas = voxel Hounsfield Unit / -1000) and expressed as percent of total lung mass. The distribution of lung aeration was assessed in 10 regions of interest (ROI) along the z-axis. Mixed effects analysis was used to evaluate the effects of PEEP and CWC.

Résultats & Discussion

At baseline, the mean P/F ratio was 113 ± 52 mmHg, consistent with severe ARDS. Respiratory system driving pressures were significantly reduced by CWC at P30 (15.9 ± 2.9 vs. 19.3 ± 3.1 cmH2O at Baseline15, P<0.01; vs. 19.1 ± 4.9 cmH2O at Baseline5, P=0.03). This effect was diminished at P60 (P=0.03 vs. Baseline15, P=ns vs. Baseline5) and P90 (P=ns vs. Baseline5 and Baseline15). Compared to Baseline15, P30 significantly reduced the end-expiratory lung mass exposed to Fgas>0.85, a common definition of hyperinflation (0.4 ± 0.3% vs. 1.5 ± 1.2%, P=0.04; Figure1), but did not increase the end-expiratory non-aerated lung mass (Fgas<0.1; 25.5 ± 8.6% vs. 24.1 ± 8.3%, P=ns). Compared to Baseline5, the non-aerated lung mass was significantly lower at Baseline15 (P<0.01) and P30 (P=0.04) but not at P60 or P90 (P=ns for both). P30 significantly reduced the mean voxel-wise Fgas in the mid-cephalic regions (ROI 4 to 8) vs. Baseline5 or Baseline15 (see Figure2). In these regions, mean Fgas did not differ between Baseline5 and P60 or P90.

Conclusion

At the lowest tested level, CWC paradoxically reduced respiratory system driving pressure by decreasing hyperinflated lung mass, predominantly in the mid-cephalic lung, without compromising the beneficial effects of high PEEP on lung recruitment. This effect was mitigated at higher levels of CWC, suggesting a threshold effect.

Auteurs

David LAGIER (1) , David NIDDAM (1), Pauline BRIGE (1), Salah BOUSSEN (1), Sarah E. GERARD (2), Marcos F. VIDAL MELO (3) - (1)Aix Marseille Université, Marseille, France, (2)Iowa University, Iowa City, Ia, États-Unis, (3)Univ Texas Medical Branch, Galveston, Tx, États-Unis

Orateur(s)

David LAGIER (Marseille)