Central venous pressure swing outperforms diaphragm ultrasound as a measure of inspiratory effort during pressure support ventilation in COVID-19 patients

Abstract

Purpose: The COVID-19-related shortage of ICU beds magnified the need of tools to properly titrate the ventilator assistance. We investigated whether bedside-available indices such as the ultrasonographic changes in diaphragm thickening ratio (TR) and the tidal swing in central venous pressure (ΔCVP) are reliable estimates of inspiratory effort, assessed as the tidal swing in esophageal pressure (ΔPes).

Methods: Prospective, observational clinical investigation in the intensive care unit of a tertiary care Hospital. Fourteen critically-ill patients were enrolled (age 64 ± 7 years, BMI 29 ± 4 kg/m²), after 6 [3; 9] days from onset of assisted ventilation. A three-level pressure support trial was performed, at 10 (PS10), 5 (PS5) and 0 cmH₂O (PS0). In each step, the esophageal and central venous pressure tidal swing were recorded, as well as diaphragm ultrasound.

Results: The reduction of pressure support was associated with an increased respiratory rate and a reduced tidal volume, while minute ventilation was unchanged. ΔPes significantly increased with reducing support (5 [3; 8] vs. 8 [14; 13] vs. 12 [6; 16] cmH₂O, p < 0.0001), as did the diaphragm TR (9.2 ± 6.1 vs. 17.6 ± 7.2 vs. 28.0 ± 10.0%, p < 0.0001) and the ΔCVP (4 [3; 7] vs. 8 [5; 9] vs. 10 [7; 11] cmH₂O, p < 0.0001). ΔCVP was significantly associated with ΔPes (R² = 0.810, p < 0.001), as was diaphragm TR, albeit with a lower coefficient of determination (R² = 0.399, p < 0.001).

Conclusions: In patients with COVID-19-associated respiratory failure undergoing assisted mechanical ventilation, ΔCVP is a better estimate of inspiratory effort than diaphragm ultrasound.

Keywords: COVID-19; Diaphragm ultrasound; Esophageal pressure; Pressure support ventilation; Weaning.

Fig. 1

Inspiratory effort measurements in a representative patient. The upper part of the figure shows the flow, airway pressure (Paw), esophageal pressure (Pes) and central venous pressure (CVP) in a representative patient. The negativization of intrathoracic pressure during inspiration lowers the esophageal and central venous pressures (ΔPes and ΔCVP, respectively). The lower panel shows the ultrasonographic view of the diaphragm: on the left side a B-mode scan of the diaphragm in the zone of apposition is performed. The diaphragm (d) is identified as a three-layer structure (non-echogenic central layer bordered by two echogenic layers, the peritoneum and the diaphragmatic pleurae, indicated by the white arrows). ic intercostal muscles, L lung, Li liver, S subcutaneous tissue. On the right side, the M-mode image of diaphragm thickening during inspiration is shown, allowing for the calculation of the thickening ratio

Fig. 2

Central venous pressure swing, esophageal pressure swing and diaphragm thickening ratio during the three steps of the study. The analysis on the variables recorded over the three different steps (PS 0, PS 5 and PS 10) was performed on all the patients by analysis of variance for repeated measurements, with step as a within-subject factor in case of normally-distributed variables, and the significance of the within-subject factors was corrected with the Greenhouse–Geisser method. Non-parametric variables were analyzed using Friedman test. Pairwise post-hoc multiple comparisons were carried out when appropriate. *p < 0.01 vs. PS 0; °p < 0.01 vs. PS 5

Fig. 3

Correlation of central venous pressure swing and diaphragm thickening ratio with the esophageal tidal pressure swing during the different phases of the study. Black dots represent the central venous pressure swing, while gray dots depict the diaphragm thickening ratio. The solid lines represent the linear predictions, while the dashed lines are their 95% confidence interval. The analysis was conducted on all patients by a linear, fixed-effects model for repeated measures to deal with the longitudinal structure of our data set (patients with repeated measurements over time). The extent of the association between variables was expressed as the coefficient of determination (R²)

Fig. 4

Esophageal and central venous pressure swing in patients with a high or a low central venous pressure during the three steps of the study. The analysis on the variables recorded over the three different steps (PS 0, PS 5 and PS 15) was performed on all the patients by analysis of variance for repeated measurements, with step as a within-subject factor in case of normally-distributed variables, and the significance of the within-subject factors was corrected with the Greenhouse–Geisser method. Non-parametric variables were analyzed using Friedman test. Pairwise post-hoc multiple comparisons were carried out when appropriate. *p < 0.01 vs. PS 0; °p < 0.01 vs. PS 5