RESUMO
BACKGROUND: Prone position is used in acute respiratory distress syndrome and in coronavirus disease 2019 (Covid-19) acute respiratory distress syndrome (ARDS). However, physiological mechanisms remain unclear. The aim of this study was to determine whether improved oxygenation was related to pulmonary shunt fraction (Q's/Q't), alveolar dead space (Vd/Vtalv) and ventilation/perfusion mismatch (V'A/Q'). METHODS: This was an international, prospective, observational, multicenter, cohort study, including six intensive care units in Sweden and Poland and 71 mechanically ventilated adult patients. RESULTS: Prone position increased PaO2:FiO2 after 30 min, by 78% (83-148 mm Hg). The effect persisted 120 min after return to supine (p < 0.001). The oxygenation index decreased 30 min after prone positioning by 43% (21-12 units). Q's/Q't decreased already after 30 min in the prone position by 17% (0.41-0.34). The effect persisted 120 min after return to supine (p < 0.005). Q's/Q't and PaO2:FiO2 were correlated both in prone (Beta -137) (p < 0.001) and in the supine position (Beta -270) (p < 0.001). V'A/Q' was unaffected and did not correlate to PaO2:FiO2 (p = 0.8). Vd/Vtalv increased at 120 min by 11% (0.55-0.61) (p < 0.05) and did not correlate to PaO2:FiO2 (p = 0.3). The ventilatory ratio increased after 30 min in the prone position by 58% (1.9-3.0) (p < 0.001). PaO2:FiO2 at baseline predicted PaO2:FiO2 at 30 min after proning (Beta 1.3) (p < 0.001). CONCLUSIONS: Improved oxygenation by prone positioning in COVID-19 ARDS patients was primarily associated with a decrease in pulmonary shunt fraction. Dead space remained high and the global V'A/Q' measure could not explain the differences in gas exchange.