Dataset Information

Machine learning predicts mortality based on analysis of ventilation parameters of critically ill patients: multi-centre validation.

ABSTRACT:

Background

Mechanical Ventilation (MV) is a complex and central treatment process in the care of critically ill patients. It influences acid-base balance and can also cause prognostically relevant biotrauma by generating forces and liberating reactive oxygen species, negatively affecting outcomes. In this work we evaluate the use of a Recurrent Neural Network (RNN) modelling to predict outcomes of mechanically ventilated patients, using standard mechanical ventilation parameters.

Methods

We performed our analysis on VENTILA dataset, an observational, prospective, international, multi-centre study, performed to investigate the effect of baseline characteristics and management changes over time on the all-cause mortality rate in mechanically ventilated patients in ICU. Our cohort includes 12,596 adult patients older than 18, associated with 12,755 distinct admissions in ICUs across 37 countries and receiving invasive and non-invasive mechanical ventilation. We carry out four different analysis. Initially we select typical mechanical ventilation parameters and evaluate the machine learning model on both, the overall cohort and a subgroup of patients admitted with respiratory disorders. Furthermore, we carry out sensitivity analysis to evaluate whether inclusion of variables related to the function of other organs, improve the predictive performance of the model for both the overall cohort as well as the subgroup of patients with respiratory disorders.

Results

Predictive performance of RNN-based model was higher with Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) of 0.72 (± 0.01) and Average Precision (AP) of 0.57 (± 0.01) in comparison to RF and LR for the overall patient dataset. Higher predictive performance was recorded in the subgroup of patients admitted with respiratory disorders with AUC of 0.75 (± 0.02) and AP of 0.65 (± 0.03). Inclusion of function of other organs further improved the performance to AUC of 0.79 (± 0.01) and AP 0.68 (± 0.02) for the overall patient dataset and AUC of 0.79 (± 0.01) and AP 0.72 (± 0.02) for the subgroup with respiratory disorders.

Conclusion

The RNN-based model demonstrated better performance than RF and LR in patients in mechanical ventilation and its subgroup admitted with respiratory disorders. Clinical studies are needed to evaluate whether it impacts decision-making and patient outcomes.

Trial registration

NCT02731898 ( https://clinicaltrials.gov/ct2/show/NCT02731898 ), prospectively registered on April 8, 2016.

SUBMITTER: Mamandipoor B

PROVIDER: S-EPMC8102841 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Objective: There exists controversy about the pathophysiology and lung mechanics of COVID-19 associated acute respiratory distress syndrome (ARDS), because some report severe hypoxemia with preserved respiratory system mechanics, contrasting with "classic" ARDS. We performed a detailed hourly analysis of the characteristics and time course of lung mechanics and biochemical analysis of patients requiring invasive mechanical ventilation (IMV) for COVID-19-associated ARDS, comparing survivors and non-survivors. Methods: Retrospective analysis of the data stored in the ICU information system of patients admitted in our hospital ICU that required IMV due to confirmed SARS-CoV-2 pneumonia between March 5th and April 30th, 2020. We compare respiratory system mechanics and gas exchange during the first ten days of IMV, discriminating volume and pressure controlled modes, between ICU survivors and non-survivors. Results: 140 patients were included, analyzing 11 138 respiratory mechanics recordings. Global mortality was 38.6%. Multivariate analysis showed that age (OR 1.092, 95% (CI 1.014-1.176)) and need of renal replacement therapies (OR 10.15, (95% CI 1.58-65.11)) were associated with higher mortality. Previous use of Angiotensin Converting Enzyme inhibitor (ACEI)/angiotensin-receptor blockers (ARBs) also seemed to show an increased mortality (OR 4.612, (95% CI 1.19-17.84)) although this significance was lost when stratifying by age. Respiratory variables start to diverge significantly between survivors and non-survivors after the 96 to 120 hours (hs) from mechanical ventilation initiation, particularly respiratory system compliance. In non survivors, mechanical power at 24 and 96 hs was higher regardless ventilatory mode. Conclusions: In patients admitted for SARS-CoV-2 pneumonia and requiring mechanical ventilation, non survivors have different respiratory system mechanics than survivors in the first 10 days of ICU admission. We propose a checkpoint at 96-120 hs to assess patients improvement or worsening in order to consider escalating to extracorporeal therapies.

Project description:INTRODUCTION: Critical illness myopathy and/or neuropathy (CRIMYNE) is frequent in intensive care unit (ICU) patients. Although complete electrophysiological tests of peripheral nerves and muscles are essential to diagnose it, they are time-consuming, precluding extensive use in daily ICU practice. We evaluated whether a simplified electrophysiological investigation of only two nerves could be used as an alternative to complete electrophysiological tests. METHODS: In this prospective, multi-centre study, 92 ICU patients were subjected to unilateral daily measurements of the action potential amplitude of the sural and peroneal nerves (compound muscle action potential [CMAP]). After the first ten days, complete electrophysiological investigations were carried out weekly until ICU discharge or death. At hospital discharge, complete neurological and electrophysiological investigations were performed. RESULTS: Electrophysiological signs of CRIMYNE occurred in 28 patients (30.4%, 95% confidence interval [CI] 21.9% to 40.4%). A unilateral peroneal CMAP reduction of more than two standard deviations of normal value showed the best combination of sensitivity (100%) and specificity (67%) in diagnosing CRIMYNE. All patients developed the electrophysiological signs of CRIMYNE within 13 days of ICU admission. Median time from ICU admission to CRIMYNE was six days (95% CI five to nine days). In 10 patients, the amplitude of the nerve action potential dropped progressively over a median of 3.0 days, and in 18 patients it dropped abruptly within 24 hours. Multi-organ failure occurred in 21 patients (22.8%, 95% CI 15.4% to 32.4%) and was strongly associated with CRIMYNE (odds ratio 4.58, 95% CI 1.64 to 12.81). Six patients with CRIMYNE died: three in the ICU and three after ICU discharge. Hospital mortality was similar in patients with and without CRIMYNE (21.4% and 17.2%; p = 0.771). At ICU discharge, electrophysiological signs of CRIMYNE persisted in 18 (64.3%) of 28 patients. At hospital discharge, diagnoses in the 15 survivors were critical illness myopathy (CIM) in six cases, critical illness polyneuropathy (CIP) in four, combined CIP and CIM in three, and undetermined in two. CONCLUSION: A peroneal CMAP reduction below two standard deviations of normal value accurately identifies patients with CRIMYNE. These should have full neurological and neurophysiological evaluations before discharge from the acute hospital.

Project description:BackgroundThe incidence of coagulopathy, which was responsible for poor outcomes, was commonly seen among patients with sepsis. In the current study, we aim to determine whether the presence of sepsis-associated coagulopathy (SAC) predicts the clinical outcomes among critically ill patients with postoperative sepsis.MethodsWe conducted a single-center retrospective cohort study by including patients with sepsis admitted to surgical ICU of Chinese PLA General Hospital from January 1, 2014 to December 31, 2018. Baseline characteristics and clinical outcomes were compared with respect to the presence of SAC. Kaplan-Meier analysis was applied to calculate survival rate, and Log-rank test was carried out to compare the differences between two groups. Furthermore, multivariable Cox and logistic and linear regression analysis were performed to assess the relationship between SAC and clinical outcomes, including hospital mortality, development of septic shock, and length of hospital stay (LOS), respectively. Additionally, both sensitivity and subgroup analyses were performed to further testify the robustness of our findings.ResultsA total of 175 patients were included in the current study. Among all included patients, 41.1% (72/175) ICU patients were identified as having SAC. In-hospital mortality rates were significantly higher in the SAC group when compared to that of the No SAC group (37.5% vs. 11.7%; p < 0.001). By performing univariable and multivariable regression analyses, presence of SAC was demonstrated to significantly correlate with an increased in-hospital mortality for patients with sepsis in surgical ICU [Hazard ratio (HR), 3.75; 95% Confidence interval (CI), 1.90-7.40; p < 0.001]. Meanwhile, a complication of SAC was found to be the independent predictor of the development of septic shock [Odds ratio (OR), 4.11; 95% CI, 1.81-9.32; p = 0.001], whereas it was not significantly associated with prolonged hospital LOS (OR, 0.97; 95% CI, 0.83-1.14; p = 0.743).ConclusionThe presence of SAC was significantly associated with increased risk of in-hospital death and septic shock among postoperative patients with sepsis admitted to ICU. Moreover, there was no statistical difference of hospital LOS between the SAC and no SAC groups.