Expiratory muscle activity and the physiological response to PEEP in hypoxemic patients

Expiratory muscle activity raises end-expiratory pleural pressure and may oppose the effects of PEEP. Whether the regional response to PEEP differs between hypoxemic patients with and without active expiration is unclear. Secondary analysis of two randomized crossover physiological studies. Thirty-one hypoxemic patients (18 invasive ventilation, 13 noninvasive support) received low (2–5 cmH 2 O) and high (14–15 cmH 2 O) PEEP in randomized order at constant inspiratory assistance. Active expiration at high PEEP was identified by gastric manometry; inspiratory effort and transpulmonary pressure by esophageal manometry; tidal ventilation (arbitrary units, a.u.) and end-expiratory lung impedance (EELI) by electrical impedance tomography. All comparisons were exploratory and unadjusted for multiplicity. Active expiration was detected in 10 (32%) patients at low PEEP and 22 (71%) at high PEEP. The PEEP-induced increase in inspiratory effort was similar between groups ( p = 0.49), but was accompanied by a rise in transdiaphragmatic pressure only with passive expiration (+3 cmH 2 O [1–4], p = 0.01), whereas with active expiration it coincided with intensified expiratory gastric pressure swings (+3 cmH 2 O [1–5], p = 0.001). End-expiratory esophageal and transpulmonary pressures rose slightly more (+4 vs + 3 cmH₂O) and less (+6 vs + 7 cmH₂O) with active expiration, without significant between-group differences. The PEEP-induced increase in global EELI was smaller with active expiration (+941 a.u. [384–1712] vs + 2348 [1215–3402], p = 0.03). With passive expiration, PEEP increased both ventral and dorsal EELI ( p = 0.004 and p = 0.008); with active expiration, only ventral EELI increased ( p = 0.04; dorsal p = 0.30). With active expiration, PEEP redistributed tidal ventilation dorsally ( p = 0.001) through a greater tidal recruitment/derecruitment fraction (13% vs 6%, p = 0.004), increasing dorsal dynamic strain (from 0.85 [0.62–1.05] to 1.10 [0.85–1.40], p = 0.001), corrected minute ventilation ( p = 0.002) and PaO₂/FiO₂ ( p = 0.001). In hypoxemic patients receiving assisted support, active expiration was common and more frequent at high PEEP, possibly to assist tidal volume generation through pre-inspiratory abdominal relaxation. With active expiration, PEEP improved oxygenation despite a smaller gain in lung volume, attenuated in the dependent lung, dorsal tidal recruitment and higher regional strain, so that oxygenation may mask injurious regional mechanics. Esophageal and transpulmonary pressure changes were directionally consistent but non-significant, suggesting a regional, predominantly dorsal effect.