Reliability and reproducibility of clinical phenotypes developed during the first wave of COVID-19: A validation study in critically ill patients from the second and third wave (preprint)

Background: During the first wave of the COVID-19 pandemic, different clinical phenotypes were published. However, none of them have been validated in subsequent waves, so their current validity is unknown. The aim of the study is to validate the unsupervised cluster model developed during the first pandemic wave in a cohort of critically ill patients from the second and third pandemic waves. Methods: Retrospective, multicentre, observational study of critically ill patients with confirmed COVID-19 disease and acute respiratory failure admitted from 74 Intensive Care Units (ICU) in Spain. To validate our original phenotypes model, we assigned a phenotype to each patient of the validation cohort using the same medoids, the same number of clusters (n= 3), the same number of variables (n= 25) and the same discretisation used in the development cohort. The performance of the classification was determined by Silhouette analysis and general linear modelling. The prognostic models were validated, and their performance was measured using accuracy test and area under curve (AUC)ROC. Results: The database included a total of 2,033 patients (mean age 63[53-92] years, 1643(70.5%) male, median APACHE II score (12[9-16]) and SOFA score (4[3-6]) points. The ICU mortality rate was 27.2%. Although the application of unsupervised cluster analysis classified patients in the validation population into 3 clinical phenotypes. Phenotype A (n=1,206 patients, 59.3%), phenotype B (n=618 patients, 30.4%) and phenotype C (n=506 patients, 24.3%), the characteristics of patients within each phenotype were significantly different from the original population. Furthermore, the silhouette coefficients were close to or below zero and the inclusion of phenotype classification in a regression model did not improve the model performance (accuracy =0.78, AUC=0.78) with respect to a standard model (accuracy = 0.79, AUC=0.79) or even worsened when the model was applied to patients within each phenotype (accuracy = 0.80, AUC 0.77 for Phenotype A, accuracy=0.73, AUC= 0.67 for phenotype B and accuracy= 0.66 , AUC= 0.76 for phenotype C ) Conclusion:  Models developed using machine learning techniques during the first pandemic wave cannot be applied with adequate performance to patients admitted in subsequent waves without prior validation. Trial Registration: The study was retrospectively registered (NCT 04948242) on June 30, 2021

Timing of Intubation and ICU Mortality in Covid-19 Patients: A Preplanned Retrospective Analysis of 4198 Critically Ill Patients (preprint)

BackgroundOptimal time to intubate patients with SARS-CoV-2 pneumonia is controversial. Whereas some authors recommend trying noninvasive respiratory support before intubate, others argue that delaying intubation can cause patient-self-induced lung injury and worsen the prognosis. We hypothesized that delayed intubation would increase the risk mortality in COVID-19 patients.MethodsThis preplanned retrospective observational study used prospectively collected data from adult patients with COVID-19 and respiratory failure admitted to 73 intensive care units between February 2020 and March 2021. Patients with limitations on life support and those with missing data were excluded.We collected demographic, laboratory, clinical variables and outcomes.Intubation was classified as 1) Very early: before or at ICU admission; 2) Early: < 24 hours after ICU admission; or 3) Late: ≥24 hours after ICU admission. We compared the early group versus those intubated late, using chi-square tests for categorical variables and the Mann-Whitney U for continuous variables. To assess the relationship between early versus late intubation and mortality, we used multivariable binary logistic regression. Statistical significance was set at p<0.05.Results We included 4198 patients [median age, 63 (54‒71) years; 70.8% male; median SOFA score, 4 (3‒7); median APACHE score, 13 (10‒18)], and median PaO2/FiO2, 131 (100‒190)]; intubation was very early in 2024 (48.2%) patients, early in 928 (22.1%), and late in 441 (10.5%). ICU mortality was 30.2% and median ICU stay was 14 (7‒28) days. Although patients in the late group were younger [62 vs. 64, respectively, p<0.05] and had less severe disease [APACHE II (13 vs. 14, respectively, p<0.05) and SOFA (3 vs. 4, respectively, p<0.05) scores], and higher PaO2/FiO2 at admission (116 vs. 100, respectively, p<0.05), mortality was higher in the late group than in the early group (36.9% vs. 31.6%, p<0.05). Late intubation was independently associated with mortality (OR1.83; 95%CI 1.35‒2.47).ConclusionsDelaying intubation beyond the first 24 hours of admission in patients with COVID-19 pneumonia increases the risk of mortality. Trial registration: The study was retrospectively registered at Clinical-Trials.gov (NCT 04948242) on the 30th June 2021.

Deploying unsupervised clustering analysis to derive clinical phenotypes and risk factors associated with mortality risk in 2,022 critically ill patients with COVID-19 in Spain (preprint)

Background: The identification of factors associated with Intensive Care Unit (ICU) mortality and derived clinical phenotypes in COVID-19 patients could help for a more tailored approach to clinical decision-making that improves prognostic outcomes. Methods: Prospective, multicenter, observational study of critically ill patients with confirmed COVID-19 disease and acute respiratory failure admitted from 63 Intensive Care Units(ICU) in Spain. The objective was to utilize an unsupervised clustering analysis to derive clinical COVID-19 phenotypes and to analyze patient’s factors associated with mortality risk. Patient features including demographics and clinical data at ICU admission were analyzed. Generalized linear models were used to determine ICU morality risk factors. The prognostic models were validated and their performance was measured using accuracy test, sensitivity, specificity and ROC curves. Results: : The database included a total of 2,022 patients (mean age 64[IQR5-71] years, 1423(70.4%) male, median APACHE II score (13[IQR10-17]) and SOFA score (5[IQR3-7]) points. The ICU mortality rate was 32.6%. Of the 3 derived phenotypes, the A(mild) phenotype (537;26.7%) included older age (<65 years), fewer abnormal laboratory values and less development of complications, B (moderate) phenotype (623,30.8%) had similar characteristics of A phenotype but were more likely to present shock. The C(severe) phenotype was the most common (857;42.5%) and was characterized by the interplay of older age (>65 years), high severity of illness and a higher likelihood of development shock. Crude ICU mortality was 20.3%, 25% and 45.4% for A, B and C phenotype respectively. The ICU mortality risk factors and model performance differed between whole population and phenotype classifications. Conclusion: The presented machine learning model identified three clinical phenotypes that significantly correlated with host-response patterns and ICU mortality. Different risk factors across the whole population and clinical phenotypes were observed which may limit the application of a “one-size-fits-all” model in practice .

Deploying Machine Learning to Validate Clinical Phenotypes and Factors Associated with Mortality Risk in 2,022 Critically Ill Patients with COVID-19 in Spain (preprint)

Background: The identification of factors associated with Intensive Care Unit (ICU) mortality and derived clinical phenotypes in COVID-19 patients could help for a more tailored approach to clinical decision-making that improves prognostic outcomes. The objective was to analyze patient’s factors associated with mortality risk and utilize a Machine Learning(ML) to derive clinical COVID-19 phenotypes.Methods: Prospective, multicenter, observational study of critically ill patients with confirmed COVID-19 disease and acute respiratory failure admitted from 63 Intensive Care Units(ICU) in Spain. Patient features including demographics and clinical data at ICU admission were analyzed. Generalized linear models were used to determine ICU morality risk factors. An unsupervised clustering analysis was applied to determine presence of phenotypes. The prognostic models were validated and their performance was measured using accuracy test, sensitivity, specificity and ROC curves.Findings: The database included a total of 2,022 patients (mean age 64[IQR5-71] years, 1423(70·4%) male, median APACHE II score (13[IQR10-17]) and SOFA score (5[IQR3-7]) points. The ICU mortality rate was 32·6%. Of the 3 derived phenotypes, the C(severe) phenotype was the most common (857;42·5%) and was characterized by the interplay of older age (>65 years), high severity of illness and a higher likelihood of development shock. The A(mild) phenotype (537;26·7%) included older age (>65 years), fewer abnormal laboratory values and less development of complications and B (moderate) phenotype (623,30·8%) had similar characteristics of A phenotype but were more likely to present shock. Crude ICU mortality was 45·4%, 25·0% and 20·3% for the C, B and A phenotype respectively. The ICU mortality risk factors and model performance differed between whole population and phenotype classifications.Interpretation: The presented ML model identified three clinical phenotypes that significantly correlated with host-response patterns and ICU mortality. Different risk factors across the whole population and clinical phenotypes were observed which may limit the application of a “one-size-fits-all” model in practice.Funding Statement: This study was supported by the Spanish Intensive Care Society(SEMICYUC) and Ricardo Barri Casanovas Foundation.Declaration of Interests: All authors declare that they have no conflicts of interest.Ethics Approval Statement: The study was approved by the reference institutional review board at Joan XXIII University Hospital (IRB# CEIM/066/2020) and each participating site with a waiver of informed consent. All data values were anonymized prior to the phenotyping which consisted of clustering clinical variables on their association with COVID-19 mortality.