Abstract Rationale Higher physiologic dead space (Vd/Vt) is associated with higher mortality in acute respiratory distress syndrome (ARDS). However, the relationship between the components of physiologic dead space, alveolar and airway dead space, and mortality has not been well examined and it is not clear how positive end expiratory pressure (PEEP) management may alter dead space. Our primary aim was to determine whether Vd/Vt and its components are associated with mortality. Additionally, we aimed to assess whether PEEP management discordant from the ARDSNet lower PEEP/FiO2 table is associated with dead space and its components. Methods This was a secondary analysis of a single-center clinical trial of a lung-protective ventilation strategy in PARDS (REDvent, 2017-2024, R01HL134666). Median physiologic, airway, and alveolar dead space within the first 72 hours of invasive mechanical ventilation were obtained from volumetric capnography data paired to clinical blood gases. Airway dead space was measured as Vdaw/kg and Vdaw/Vte. PEEP discordance was defined as the difference between clinical set PEEP and the ARDSNet low PEEP/FiO2 table recommendation. The primary outcome was mortality, with a secondary outcome of 28-day ventilator-free days (VFD). Results Among 183 patients, 24 (13%) died. Based on 72-hour medians, 35 (19%) had high airway dead space alone (≥3 ml/kg), 31 (17%) had high alveolar dead space (Vdalv/Vtalv) (≥0.2), 43 (24%) had both, and 74 (40%) had neither. Vdalv/Vtalv (p 0.001), Vdaw/kg (p = 0.004), Vdaw/Vte (p 0.001), and Vd/Vt (p 0.001) were each associated with higher mortality in univariate modeling. Vdaw/kg (p = 0.038), Vdaw/Vte (p 0.001), Vd/Vt (p = 0.002) remained associated with mortality after adjustment for OI (Figure 1). Vdalv/Vtalv (p = 0.03), Vdaw/Vte (p = 0.002), and Vd/Vt (p 0.001) were associated with fewer VFDs, with Vdaw/Vte (p = 0.009) and Vd/Vt (p = 0.029) remaining associated after adjustment for confounding variables (Figure 1). A PEEP 4 cmH2O or greater above the ARDSNet recommendation was associated with lower Vdalv/Vtalv (p = 0.048) and higher Vdaw/kg (p = 0.036), while PEEP 4 cmH2O or greater below the ARDSNet recommendation was associated with higher alveolar dead space (p 0.001) and lower Vdaw/Vte (p 0.001). Vd/Vt was not associated with PEEP discordance. Conclusions Physiologic dead space and its components are markers of mortality risk in PARDS and have differing associations with PEEP management. Airway dead space, which can be measured continuously and non-invasively with volumetric capnography, may be an important marker of mortality risk and response to PEEP management that should be investigated further. This abstract is funded by: NHLBI R01HL134666
Leeds et al. (Fri,) studied this question.