Time-Dependent Effects of Prone Position on Ventilation-Perfusion Matching Assessed by Electrical Impedance Tomography in Patients With ARDS: A Prospective Physiological Study

Abstract

Background:

Contrast-enhanced electrical impedance tomography (EIT) has emerged as a valuable tool for bedside monitoring of pulmonary ventilation and perfusion. It provides valuable information on the physiological mechanism underlying oxygenation responses to prone position (PP) in patients with ARDS. However, the impact of PP duration on the changes of ventilation and perfusion remains unclear.

Methods:

Mechanically ventilated ARDS patients who underwent PP were consented and included in this study after obtaining IRB approval (No. B2019-230(2)R). EIT evaluations and PaO₂/FIO₂ were performed at six time points during the first PP session: at supine position (SP) before PP initiation, 1 h after PP (PP₁), 3 h after PP (PP₃), 9 h after PP (PP₉), the end of PP (PP₁₆), and 3 h after returning to supine position (RE-SP₃). Repeated ANOVA tests were used to compare the differences in the variables at six time points.

Results:

Eighteen patients (17 males) with a mean age of 72 ± 9 years were enrolled in this study. When comparing PP to SP, PP led to significant improvements in oxygenation. The PaO₂/FIO₂ ratio showed a steady increase with each PP session (all P < .05), peaking at PP₁₆. Although PaO₂/FIO₂ decreased at RE-SP₃, it remained higher than that at SP (163.8 ± 86.5 vs 110.7 ± 37.8 mm Hg, P = .003) (Fig 1). Compared to SP, dorsal ventilation significantly increased at PP₁ (28.5 ± 9.2 vs 38.4 ± 12.0%, P = .047) while ventral ventilation significantly decreased at PP₁ (71.5 ± 9.2 vs 61.6 ± 12.0%, P = .047) (Fig 2A), and they consistently changed as the PP duration increased, with more dorsal ventilation at PP₁₆ than PP₁ (50.3 ± 14.1 vs 38.4 ± 12.0%, P < .001) and less ventral ventilation at PP₁₆ than PP₁ (49.7 ± 14.1 vs 61.6 ± 12.0%, P = .021). Dorsal and ventral perfusion remained stable during the first three hours of PP, but dorsal perfusion started to increase and ventral perfusion started to decrease significantly at PP₁₆ (Fig 2B). Compared to SP, there was a significant increase in global V/Q matching at PP₃ (54.1 ± 17.5 vs 64.7 ± 13.9%, P = .008) (Fig 2C), primarily occurred in dorsal area (Fig 2D).

Conclusions:

Early improvement of oxygenation and global V/Q matching observed during PP was primarily attributed to the immediate and early alterations in ventilation distribution, which persisted throughout the middle and late stages. However, changes in perfusion were only evident during the middle and late stage of PP, contributing to further improvements in oxygenation and global V/Q matching.

Evolution of PaO2/FiO2 ratios at six time points during the first PP session: at supine position (SP) before PP initiation, 1h after PP(PP₁), 3h after PP (PP₃), 9h after PP(PP₉), the end of PP(PP₁₆), and 3h after returning to supine position (RE-SP₃).

Comparisons of Ventilation distribution(%) (A), Perfusion distribution (%) (B) at SP, PP₁，PP₃, PP₉, PP₁₆ and RE-SP₃. Evolution of global V/Q matching (%), Shunt-EIT (%), and Dead space-EIT (%) at SP, PP₁, PP₃, PP₉, PP₁₆ and RE-SP₃ (C). Evolution of V/Q matching (%), Shunt-EIT (%), and Dead space-EIT (%) at SP, PP₁, PP₃, PP₉, PP₁₆ and RE-SP₃ in dorsal and ventral region at SP, PP₁, PP₃, PP₉, PP₁₆ and RE-SP₃ (D, E).