Diminishing Efficacy of Prone Positioning With Late Application in Evolving Lung Injury

Deep learning segmentation and registration-driven lung parenchymal volume and movement CT analysis in prone positioning

PURPOSE

To conduct a volumetric and movement analysis of lung parenchyma in prone positioning using deep neural networks (DNNs).

METHOD

We included patients with suspected interstitial lung abnormalities or disease who underwent full-inspiratory supine and prone chest CT at a single institution between June 2021 and March 2022. A thoracic radiologist visually assessed the fibrosis extent in the total lung (using units of 10%) on supine CT. After preprocessing the images into 192×192×192 resolution, a DNN automatically segmented the whole lung and pulmonary lobes in prone and supine CT images. Affine registration matched the patient's center and location, and the DNN deformably registered prone and supine CT images to calculate the x-, y-, z-axis, and 3D pixel movements.

RESULTS

In total, 108 CT pairs had successful registration. Prone positioning significantly increased the left lower (90.2±69.5 mL, P = 0.000) and right lower lobar volumes (52.5±74.2 mL, P = 0.000). During deformable registration, the average maximum whole-lung pixel movements between the two positions were 1.5, 1.9, 1.6, and 2.8 cm in each axis and 3D plane. Compared to patients with <30% fibrosis, those with ≥30% fibrosis had smaller volume changes (P<0.001) and smaller pixel movements in all axes between the positions (P = 0.000-0.007). Forced vital capacity (FVC) correlated with the left lower lobar volume increase (Spearman correlation coefficient, 0.238) and the maximum whole-lung pixel movements in all axes (coefficients, 0.311 to 0.357).

CONCLUSIONS

Prone positioning led to the preferential expansion of the lower lobes, correlated with FVC, and lung fibrosis limited lung expansion during prone positioning.

Electrical Impedance Tomography Identifies Evolution of Regional Perfusion in a Porcine Model of Acute Respiratory Distress Syndrome

BACKGROUND

Bedside electrical impedance tomography could be useful to visualize evolving pulmonary perfusion distributions when acute respiratory distress syndrome worsens or in response to ventilatory and positional therapies. In experimental acute respiratory distress syndrome, this study evaluated the agreement of electrical impedance tomography and dynamic contrast-enhanced computed tomography perfusion distributions at two injury time points and in response to increased positive end-expiratory pressure (PEEP) and prone position.

METHODS

Eleven mechanically ventilated (VT 8 ml · kg-1) Yorkshire pigs (five male, six female) received bronchial hydrochloric acid (3.5 ml · kg-1) to invoke lung injury. Electrical impedance tomography and computed tomography perfusion images were obtained at 2 h (early injury) and 24 h (late injury) after injury in supine position with PEEP 5 and 10 cm H2O. In eight animals, electrical impedance tomography and computed tomography perfusion imaging were also conducted in the prone position. Electrical impedance tomography perfusion (QEIT) and computed tomography perfusion (QCT) values (as percentages of image total) were compared in eight vertical regions across injury stages, levels of PEEP, and body positions using mixed-effects linear regression. The primary outcome was agreement between QEIT and QCT, defined using limits of agreement and Pearson correlation coefficient.

RESULTS

Pao2/Fio2 decreased over the course of the experiment (healthy to early injury, -253 [95% CI, -317 to -189]; early to late injury, -88 [95% CI, -151 to -24]). The limits of agreement between QEIT and QCT were -4.66% and 4.73% for the middle 50% quantile of average regional perfusion, and the correlation coefficient was 0.88 (95% CI, 0.86 to 0.90]; P < 0.001). Electrical impedance tomography and computed tomography showed similar perfusion redistributions over injury stages and in response to increased PEEP. QEIT redistributions after positional therapy underestimated QCT in ventral regions and overestimated QCT in dorsal regions.

CONCLUSIONS

Electrical impedance tomography closely approximated computed tomography perfusion measures in experimental acute respiratory distress syndrome, in the supine position, over injury progression and with increased PEEP. Further validation is needed to determine the accuracy of electrical impedance tomography in measuring perfusion redistributions after positional changes.

EDITOR’S PERSPECTIVE

Effects of early versus delayed application of prone position on ventilation-perfusion mismatch in patients with acute respiratory distress syndrome: a prospective observational study

BACKGROUND

Prone position has been shown to improve oxygenation and survival in patients with early acute respiratory distress syndrome (ARDS). These beneficial effects are partly mediated by improved ventilation/perfusion (V/Q) distribution. Few studies have investigated the impact of early versus delayed proning on V/Q distribution in patients with ARDS. The aim of this study was to assess the regional ventilation and perfusion distribution in early versus persistent ARDS after prone position.

METHODS

This is a prospective, observational study from June 30, 2021, to October 1, 2022 at the medical ICU in Zhongda Hospital, Southeast University. Fifty-seven consecutive adult patients with moderate-to-severe ARDS ventilated in supine and prone position. Electrical impedance tomography was used to study V/Q distribution in the supine position and 12 h after a prone session.

RESULTS

Of the 57 patients, 33 were early ARDS (≤ 7 days) and 24 were persistent ARDS (> 7 days). Oxygenation significantly improved after proning in early ARDS (157 [121, 191] vs. 190 [164, 245] mm Hg, p < 0.001), whereas no significant change was found in persistent ARDS patients (168 [136, 232] vs.177 [155, 232] mm Hg, p = 0.10). Compared to supine position, prone reduced V/Q mismatch in early ARDS (28.7 [24.6, 35.4] vs. 22.8 [20.0, 26.8] %, p < 0.001), but increased V/Q mismatch in persistent ARDS (23.8 [19.8, 28.6] vs. 30.3 [24.5, 33.3] %, p = 0.006). In early ARDS, proning significantly reduced shunt in the dorsal region and dead space in the ventral region. In persistent ARDS, proning increased global shunt. A significant correlation was found between duration of ARDS onset to proning and the change in V/Q distribution (r = 0.54, p < 0.001).

CONCLUSIONS

Prone position significantly reduced V/Q mismatch in patients with early ARDS, while it increased V/Q mismatch in persistent ARDS patients. Trial registration ClinicalTrials.gov (NCT05207267, principal investigator Ling Liu, date of registration 2021.08.20).

Regional lung perfusion using different indicators in electrical impedance tomography

Management of acute respiratory distress syndrome (ARDS) is classically guided by protecting the injured lung and mitigating damage from mechanical ventilation. Yet the natural history of ARDS is also dictated by disruption in lung perfusion. Unfortunately, diagnosis and treatment are hampered by the lack of bedside perfusion monitoring. Electrical impedance tomography is a portable imaging technique that can estimate regional lung perfusion in experimental settings from the kinetic analysis of a bolus of an indicator with high conductivity. Hypertonic sodium chloride has been the standard indicator. However, hypertonic sodium chloride is often inaccessible in the hospital, limiting practical adoption. We investigated whether regional lung perfusion measured using electrical impedance tomography is comparable between indicators. Using a swine lung injury model, we determined regional lung perfusion (% of total perfusion) in five pigs, comparing 12% sodium chloride to 8.4% sodium bicarbonate across stages of lung injury and experimental conditions (body position, positive end-expiratory pressure). Regional lung perfusion for four lung regions was determined from maximum slope analysis of the indicator-based impedance signal. Estimates of regional lung perfusion between indicators were compared in the lung overall and within four lung regions. Regional lung perfusion estimated with a sodium bicarbonate indicator agreed with a hypertonic sodium chloride indicator overall (mean bias 0%, limits of agreement -8.43%, 8.43%) and within lung quadrants. The difference in regional lung perfusion between indicators did not change across experimental conditions. Sodium bicarbonate may be a comparable indicator to estimate regional lung perfusion using electrical impedance tomography.NEW & NOTEWORTHY Electrical impedance tomography is an emerging tool to measure regional lung perfusion using kinetic analysis of a conductive indicator. Hypertonic sodium chloride is the standard agent used. We measured regional lung perfusion using another indicator, comparing hypertonic sodium chloride to sodium bicarbonate in an experimental swine lung injury model. We found strong agreement between the two indicators. Sodium bicarbonate may be a comparable indicator to measure regional lung perfusion with electrical impedance tomography.

Improving pulmonary perfusion assessment by dynamic contrast-enhanced computed tomography in an experimental lung injury model

Pulmonary perfusion has been poorly characterized in acute respiratory distress syndrome (ARDS). Optimizing protocols to measure pulmonary blood flow (PBF) via dynamic contrast-enhanced (DCE) computed tomography (CT) could improve understanding of how ARDS alters pulmonary perfusion. In this study, comparative evaluations of injection protocols and tracer-kinetic analysis models were performed based on DCE-CT data measured in ventilated pigs with and without lung injury. Ten Yorkshire pigs (five with lung injury, five healthy) were anesthetized, intubated, and mechanically ventilated; lung injury was induced by bronchial hydrochloric acid instillation. Each DCE-CT scan was obtained during a 30-s end-expiratory breath-hold. Reproducibility of PBF measurements was evaluated in three pigs. In eight pigs, undiluted and diluted Isovue-370 were separately injected to evaluate the effect of contrast viscosity on estimated PBF values. PBF was estimated with the peak-enhancement and the steepest-slope approach. Total-lung PBF was estimated in two healthy pigs to compare with cardiac output measured invasively by thermodilution in the pulmonary artery. Repeated measurements in the same animals yielded a good reproducibility of computed PBF maps. Injecting diluted isovue-370 resulted in smaller contrast-time curves in the pulmonary artery (P < 0.01) and vein (P < 0.01) without substantially diminishing peak signal intensity (P = 0.46 in the pulmonary artery) compared with the pure contrast agent since its viscosity is closer to that of blood. As compared with the peak-enhancement model, PBF values estimated by the steepest-slope model with diluted contrast were much closer to the cardiac output (R2 = 0.82) as compared with the peak-enhancement model. DCE-CT using the steepest-slope model and diluted contrast agent provided reliable quantitative estimates of PBF.NEW & NOTEWORTHY Dynamic contrast-enhanced CT using a lower-viscosity contrast agent in combination with tracer-kinetic analysis by the steepest-slope model improves pulmonary blood flow measurements and assessment of regional distributions of lung perfusion.

Effect of convalescent plasma transfusion on outcomes of coronavirus disease 2019: a meta-analysis with trial sequential analysis

The aim of this review was to update evidence for benefit of convalescent plasma transfusion (CPT) in patients with coronavirus disease 2019 (COVID-19). Databases were searched for randomized controlled trials (RCT) comparing CPT plus standard treatment versus standard treatment only in adults with COVID-19. Primary outcome measures were mortality and need for invasive mechanical ventilation (IMV). Twenty-Six RCT involving 19,816 patients were included in meta-analysis for mortality. Quantitative synthesis showed no statistically significant benefit of adding CPT to standard treatment (RR = 0.97, 95% CI = 0.92 to 1.02) with unimportant heterogeneity (Q(25) = 26.48, p = .38, I² = 0.00%). Trim-and-fill-adjusted effect size was unimportantly changed and level of evidence was graded as high. Trial sequential analysis (TSA) indicated information size was adequate and CPT was futile. Seventeen trials involving 16,083 patients were included in meta-analysis for need of IMV. There was no statistically significant effect of CPT (RR = 1.02, 95% CI = 0.95 to 1.10) with unimportant heterogeneity (Q(16) = 9.43, p = .89, I² = 3.30%). Trim-and-fill-adjusted effect size was trivially changed and level of evidence was graded as high. TSA showed information size was adequate and indicated futility of CPT. It is concluded with high level of certainty that CPT added to standard treatment of COVID-19 is not associated with reduced mortality or need of IMV compared with standard treatment alone. In view of these findings, further trials on efficacy of CPT in COVID-19 patients are probably not needed.

Timing of Prone Positioning During Venovenous Extracorporeal Membrane Oxygenation for Acute Respiratory Distress Syndrome

OBJECTIVES

To assess the association of timing to prone positioning (PP) during venovenous extracorporeal membrane oxygenation (V-V ECMO) with the probability of being discharged alive from the ICU at 90 days (primary endpoint) and the improvement of the respiratory system compliance (Cpl,rs).

DESIGN

Pooled individual data analysis from five original observational cohort studies.

SETTING

European extracorporeal membrane oxygenation (ECMO) centers.

PATIENTS

Acute respiratory distress syndrome (ARDS) patients who underwent PP during ECMO.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Time to PP during V-V ECMO was explored both as a continuous and a categorical variable with Cox proportional hazard models. Three hundred patients were included in the analysis. The longer the time to PP during V-V ECMO, the lower the adjusted probability of alive ICU discharge (adjusted hazard ratio [HR] 0.90 for each day increase; 95% CI, 0.87-0.93). Two hundred twenty-three and 77 patients were included in the early PP (≤ 5 d) and late PP (> 5 d) groups, respectively. The cumulative 90-day probability of being discharged alive from the ICU was 61% in the early PP group vs 36% in the late PP group (log-rank test, p <0.001). This benefit was maintained after adjustment for confounders (adjusted HR, 2.52; 95% CI, 1.66-3.81; p <0.001). In the early PP group, PP was associated with a significant improvement of Cpl,rs (4 ± 9 mL/cm H2O vs 0 ± 12 in the late PP group, p=0.038).

CONCLUSIONS

In a large cohort of ARDS patients on ECMO, early PP during ECMO was associated with a higher probability of being discharged alive from the ICU at 90 days and a greater improvement of Cpl,rs.

Prone Positioning and Neuromuscular Blocking Agents as Adjunctive Therapies in Mechanically Ventilated Patients with Acute Respiratory Distress Syndrome

Neuromuscular blocking agents (NMBAs) and prone position (PP) are two major adjunctive therapies that can improve outcome in moderate-to-severe acute respiratory distress syndrome. NMBA should be used once lung-protective mechanical ventilation has been set, for 48 hours or less and as a continuous intravenous infusion. PP should be used as early as possible for long sessions; in COVID-19 its use has exploded. In nonintubated patients, PP might reduce the rate of intubation but not mortality. The goal of this article is to perform a narrative review on the pathophysiological rationale, the clinical effects, and the clinical use and recommendations of both NMBA and PP.

Electrical Impedance Tomography in Acute Respiratory Distress Syndrome Management

OBJECTIVE

To describe, through a narrative review, the physiologic principles underlying electrical impedance tomography, and its potential applications in managing acute respiratory distress syndrome (ARDS). To address the current evidence supporting its use in different clinical scenarios along the ARDS management continuum.

DATA SOURCES

We performed an online search in Pubmed to review articles. We searched MEDLINE, Cochrane Central Register, and clinicaltrials.gov for controlled trials databases.

STUDY SELECTION

Selected publications included case series, pilot-physiologic studies, observational cohorts, and randomized controlled trials. To describe the rationale underlying physiologic principles, we included experimental studies.

DATA EXTRACTION

Data from relevant publications were reviewed, analyzed, and its content summarized.

DATA SYNTHESIS

Electrical impedance tomography is an imaging technique that has aided in understanding the mechanisms underlying multiple interventions used in ARDS management. It has the potential to monitor and predict the response to prone positioning, aid in the dosage of flow rate in high-flow nasal cannula, and guide the titration of positive-end expiratory pressure during invasive mechanical ventilation. The latter has been demonstrated to improve physiologic and mechanical parameters correlating with lung recruitment. Similarly, its use in detecting pneumothorax and harmful patient-ventilator interactions such as pendelluft has been proven effective. Nonetheless, its impact on clinically meaningful outcomes remains to be determined.

CONCLUSIONS

Electrical impedance tomography is a potential tool for the individualized management of ARDS throughout its different stages. Clinical trials should aim to determine whether a specific approach can improve clinical outcomes in ARDS management.

Comparing Prone Positioning Use in COVID-19 Versus Historic Acute Respiratory Distress Syndrome

Use of prone positioning in patients with acute respiratory distress syndrome (ARDS) from COVID-19 may be greater than in patients treated for ARDS before the pandemic. However, the magnitude of this increase, sources of practice variation, and the extent to which use adheres to guidelines is unknown.

Early prolonged prone position in noninvasively ventilated patients with SARS-CoV-2-related moderate-to-severe hypoxemic respiratory failure: clinical outcomes and mechanisms for treatment response in the PRO-NIV study

Background

Whether prone position (PP) improves clinical outcomes in COVID-19 pneumonia treated with noninvasive ventilation (NIV) is unknown. We evaluated the effect of early PP on 28-day NIV failure, intubation and death in noninvasively ventilated patients with moderate-to-severe acute hypoxemic respiratory failure due to COVID-19 pneumonia and explored physiological mechanisms underlying treatment response.

Methods

In this controlled non-randomized trial, 81 consecutive prospectively enrolled patients with COVID-19 pneumonia and moderate-to-severe (paO2/FiO2 ratio < 200) acute hypoxemic respiratory failure treated with early PP + NIV during Dec 2020–May 2021were compared with 162 consecutive patients with COVID-19 pneumonia matched for age, mortality risk, severity of illness and paO2/FiO2 ratio at admission, treated with conventional (supine) NIV during Apr 2020–Dec 2020 at HUMANITAS Gradenigo Subintensive Care Unit, after propensity score adjustment for multiple baseline and treatment-related variables to limit confounding. Lung ultrasonography (LUS) was performed at baseline and at day 5. Ventilatory parameters, physiological dead space indices (DSIs) and circulating inflammatory and procoagulative biomarkers were monitored during the initial 7 days.

Results

In the intention-to-treat analysis. NIV failure occurred in 14 (17%) of PP patients versus 70 (43%) of controls [HR = 0.32, 95% CI 0.21–0.50; p < 0.0001]; intubation in 8 (11%) of PP patients versus 44 (30%) of controls [HR = 0.31, 95% CI 0.18–0.55; p = 0.0012], death in 10 (12%) of PP patients versus 59 (36%) of controls [HR = 0.27, 95% CI 0.17–0.44; p < 0.0001]. The effect remained significant within different categories of severity of hypoxemia (paO2/FiO2 < 100 or paO2/FiO2 100–199 at admission). Adverse events were rare and evenly distributed. Compared with controls, PP therapy was associated with improved oxygenation and DSIs, reduced global LUS severity indices largely through enhanced reaeration of dorso-lateral lung regions, and an earlier decline in inflammatory markers and D-dimer. In multivariate analysis, day 1 CO2 response outperformed O2 response as a predictor of LUS changes, NIV failure, intubation and death.

Conclusion

Early prolonged PP is safe and is associated with lower NIV failure, intubation and death rates in noninvasively ventilated patients with COVID-19-related moderate-to-severe hypoxemic respiratory failure. Early dead space reduction and reaeration of dorso-lateral lung regions predicted clinical outcomes in our study population.

Clinical trial registration

ISRCTN23016116. Retrospectively registered on May 1, 2021.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13054-022-03937-x.