Predicting responders to prone positioning in mechanically ventilated patients with COVID-19 using machine learning

Subphenotyping prone position responders with machine learning

BACKGROUND

Acute respiratory distress syndrome (ARDS) is a heterogeneous condition with varying response to prone positioning. We aimed to identify subphenotypes of ARDS patients undergoing prone positioning using machine learning and assess their association with mortality and response to prone positioning.

METHODS

In this retrospective observational study, we enrolled 353 mechanically ventilated ARDS patients who underwent at least one prone positioning cycle. Unsupervised machine learning was used to identify subphenotypes based on respiratory mechanics, oxygenation parameters, and demographic variables collected in supine position. The primary outcome was 28-day mortality. Secondary outcomes included response to prone positioning in terms of respiratory system compliance, driving pressure, PaO2/FiO2 ratio, ventilatory ratio, and mechanical power.

RESULTS

Three distinct subphenotypes were identified. Cluster 1 (22.9% of whole cohort) had a higher PaO2/FiO2 ratio and lower Positive End-Expiratory Pressure (PEEP). Cluster 2 (51.3%) had a higher proportion of COVID-19 patients, lower driving pressure, higher PEEP, and higher respiratory system compliance. Cluster 3 (25.8%) had a lower pH, higher PaCO2, and higher ventilatory ratio. Mortality differed significantly across clusters (p = 0.03), with Cluster 3 having the highest mortality (56%). There were no significant differences in the proportions of responders to prone positioning for any of the studied parameters. Transpulmonary pressure measurements in a subcohort did not improve subphenotype characterization.

CONCLUSIONS

Distinct ARDS subphenotypes with varying mortality were identified in patients undergoing prone positioning; however, predicting which patients benefited from this intervention based on available data was not possible. These findings underscore the need for continued efforts in phenotyping ARDS through multimodal data to better understand the heterogeneity of this population.

Machine learning-based derivation and validation of three immune phenotypes for risk stratification and prognosis in community-acquired pneumonia: a retrospective cohort study

Background

The clinical presentation of Community-acquired pneumonia (CAP) in hospitalized patients exhibits heterogeneity. Inflammation and immune responses play significant roles in CAP development. However, research on immunophenotypes in CAP patients is limited, with few machine learning (ML) models analyzing immune indicators.

Methods

A retrospective cohort study was conducted at Xinhua Hospital, affiliated with Shanghai Jiaotong University. Patients meeting predefined criteria were included and unsupervised clustering was used to identify phenotypes. Patients with distinct phenotypes were also compared in different outcomes. By machine learning methods, we comprehensively assess the disease severity of CAP patients.

Results

A total of 1156 CAP patients were included in this research. In the training cohort (n=809), we identified three immune phenotypes among patients: Phenotype A (42.0%), Phenotype B (40.2%), and Phenotype C (17.8%), with Phenotype C corresponding to more severe disease. Similar results can be observed in the validation cohort. The optimal prognostic model, SuperPC, achieved the highest average C-index of 0.859. For predicting CAP severity, the random forest model was highly accurate, with C-index of 0.998 and 0.794 in training and validation cohorts, respectively.

Conclusion

CAP patients can be categorized into three distinct immune phenotypes, each with prognostic relevance. Machine learning exhibits potential in predicting mortality and disease severity in CAP patients by leveraging clinical immunological data. Further external validation studies are crucial to confirm applicability.

The "ZEEP-PEEP test" to evaluate the response to positive end-expiratory pressure delivered by helmet: A prospective physiologic study

Introduction

The improvement in oxygenation after helmet application in hypoxemic patients may be explained by the alveolar recruitment obtained with positive end expiratory pressure (PEEP) or by the administration of a more accurate inspiratory fraction of oxygen (FiO2). We have designed the "ZEEP-PEEP test", capable to distinguish between the FiO2-related or PEEP-related oxygenation improvement. Our primary aim was to describe the use of this test during helmet CPAP to assess the oxygenation improvement attributable to PEEP application.

Material and methods

We performed a prospective physiological study including adult critically ill patients. Respiratory and hemodynamic parameters were recorded before helmet application (PRE step), after helmet application without PEEP (ZEEP step) and after the application of the PEEP valve (PEEP step), while maintaining a constant FiO2. We defined as "PEEP responders" patients showing a PaO2/FiO2 ratio improvement ≥10% after PEEP application.

Results

93 patients were enrolled. Compared to the PRE step, PaO2/FiO2 ratio was significantly improved during helmet CPAP both at ZEEP and PEEP step (189 ± 55, 219 ± 74 and 241 ± 82 mmHg, respectively, p < 0.01). Both PEEP responders (41%) and non-responders showed a significant improvement of PaO2/FiO2 ratio after the application of helmet at ZEEP, PEEP responders also showed a significant improvement of oxygenation after PEEP application (208 ± 70 vs 267 ± 85, p < 0.01).

Conclusions

Helmet CPAP improved oxygenation. This improvement was not only due to the PEEP effect, but also to the increase of the effective inspired FiO2. Performing the ZEEP-PEEP test may help to identify patients who benefit from PEEP.

Changes in Driving Pressure vs Oxygenation as Predictor of Mortality in Moderate to Severe Acute Respiratory Distress Syndrome Patients Receiving Prone Position Ventilation

Background

Prone position ventilation (PPV) causes improvement in oxygenation, nevertheless, mortality in severe acute respiratory distress syndrome (ARDS) remains high. The changes in the driving pressure (DP) and its role in predicting mortality in moderate to severe ARDS patients receiving PPV is unexplored.

Methods

A prospective observational study, conducted between September 2020 and February 2023 on moderate-severe ARDS patients requiring PPV. The values of DP and oxygenation (ratio of partial pressure of arterial oxygen to fraction of inspired oxygen [PaO2/FiO2]) before, during, and after PPV were recorded. The aim was to compare the DP and oxygenation before, during and after PPV sessions among moderate- severe ARDS patients, and determine the best predictor of mortality.

Results

Total of 52 patients were included; 28-day mortality was 57%. Among the survivors, DP prior to PPV as compared to post-PPV session reduced significantly, from 16.36 ± 2.57 cmH2O to 13.91 ± 1.74 cmH2O (p-value < 0.001), whereas DP did not reduce in the non-survivors (19.43 ± 3.16 to 19.70 ± 3.15 cmH2O (p-value = 0.318)]. Significant improvement in PaO2/FiO2 before PPV to post-PPV among both the survivors [92.75 [67.5-117.75]) to [205.50 (116.25-244.50)], (p-value < 0.001) and also among the non-survivors [87.90 (67.75-100.75)] to [112 (88.00-146.50)], (p-value < 0.001) was noted. Logistic regression analysis showed DP after PPV session as best predictor of mortality (p-value = 0.044) and its AUROC to predict mortality was 0.939, cut-off ≥16 cmH2O, 90% sensitivity, 82% specificity. The Kaplan-Meier curve of DP after PPV ≥16 cmH2O and <16 cmH2O was significant (Log-rank Mantel-Cox p-value < 0.001).

Conclusion

Prone position ventilation-induced decrease in DP is prognostic marker of survival than the increase in PaO2/FiO2. There is a primacy of DP, rather than oxygenation, in predicting mortality in moderate-severe ARDS. Post-PPV session DP ≥16 cmH2O was an independent predictor of mortality.

How to cite this article

Todur P, Nileshwar A, Chaudhuri S, Shanbhag V, Cherisma C. Changes in Driving Pressure vs Oxygenation as Predictor of Mortality in Moderate to Severe Acute Respiratory Distress Syndrome Patients Receiving Prone Position Ventilation. Indian J Crit Care Med 2024;28(2):134-140.

Validity of the total SOFA score in patients ≥ 80 years old acutely admitted to intensive care units: a post-hoc analysis of the VIP2 prospective, international cohort study

BACKGROUND

Little is known about the performance of the Sequential Organ Failure Assessment (SOFA) score in older critically ill adults. We aimed to evaluate the prognostic impact of physiological disturbances in the six organ systems included in the SOFA score.

METHODS

We analysed previously collected data from a prospective cohort study conducted between 2018 and 2019 in 22 countries. Consecutive patients ≥ 80 years old acutely admitted to intensive care units (ICUs) were eligible for inclusion. Patients were followed up for 30 days after admission to the ICU. We used logistic regression to study the association between increasing severity of organ dysfunction and mortality.

RESULTS

The median SOFA score among 3882 analysed patients was equal to 6 (IQR: 4-9). Mortality was equal to 26.1% (95% CI 24.7-27.5%) in the ICU and 38.7% (95% CI 37.1-40.2%) at day 30. Organ failure defined as a SOFA score ≥ 3 was associated with variable adjusted odds ratios (aORs) for ICU mortality dependant on the organ system affected: respiratory, 1.53 (95% CI 1.29-1.81); cardiovascular 1.69 (95% CI 1.43-2.01); hepatic, 1.74 (95% CI 0.97-3.15); renal, 1.87 (95% CI 1.48-2.35); central nervous system, 2.79 (95% CI 2.34-3.33); coagulation, 2.72 (95% CI 1.66-4.48). Modelling consecutive levels of organ dysfunction resulted in aORs equal to 0.57 (95% CI 0.33-1.00) when patients scored 2 points in the cardiovascular system and 1.01 (0.79-1.30) when the cardiovascular SOFA equalled 3.

CONCLUSIONS

Different components of the SOFA score have different prognostic implications for older critically ill adults. The cardiovascular component of the SOFA score requires revision.

Dead space ventilation-related indices: bedside tools to evaluate the ventilation and perfusion relationship in patients with acute respiratory distress syndrome

Cumulative evidence has demonstrated that the ventilatory ratio closely correlates with mortality in acute respiratory distress syndrome (ARDS), and a primary feature in coronavirus disease 2019 (COVID-19)-ARDS is increased dead space that has been reported recently. Thus, new attention has been given to this group of dead space ventilation-related indices, such as physiological dead space fraction, ventilatory ratio, and end-tidal-to-arterial PCO₂ ratio, which, albeit distinctive, are all global indices with which to assess the relationship between ventilation and perfusion. These parameters have already been applied to positive end expiratory pressure titration, prediction of responses to the prone position and the field of extracorporeal life support for patients suffering from ARDS. Dead space ventilation-related indices remain hampered by several deflects; notwithstanding, for this catastrophic syndrome, they may facilitate better stratifications and identifications of subphenotypes, thereby providing therapy tailored to individual needs.