Deploying unsupervised clustering analysis to derive clinical phenotypes and risk factors associated with mortality risk in 2022 critically ill patients with COVID-19 in Spain.

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 ICUs 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 2022 patients (mean age 64 [IQR 5-71] years, 1423 (70.4%) male, median APACHE II score (13 [IQR 10-17]) and SOFA score (5 [IQR 3-7]) 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.

Applicability of an unsupervised cluster model developed on first wave COVID-19 patients in second/third wave critically ill patients.

OBJECTIVE: To validate the unsupervised cluster model (USCM) developed during the first pandemic wave in a cohort of critically ill patients from the second and third pandemic waves. DESIGN: Observational, retrospective, multicentre study. SETTING: Intensive Care Unit (ICU). PATIENTS: Adult patients admitted with COVID-19 and respiratory failure during the second and third pandemic waves. INTERVENTIONS: None. MAIN VARIABLES OF INTEREST: Collected data included demographic and clinical characteristics, comorbidities, laboratory tests and ICU outcomes. To validate our original USCM, we assigned a phenotype to each patient of the validation cohort. The performance of the classification was determined by Silhouette coefficient (SC) and general linear modelling. In a post-hoc analysis we developed and validated a USCM specific to the validation set. The model's performance was measured using accuracy test and area under curve (AUC) ROC. RESULTS: A total of 2330 patients (mean age 63 [53-82] years, 1643 (70.5%) male, median APACHE II score (12 [9-16]) and SOFA score (4 [3-6]) were included. The ICU mortality was 27.2%. The USCM classified patients into 3 clinical phenotypes: A (n = 1206 patients, 51.8%); B (n = 618 patients, 26.5%), and C (n = 506 patients, 21.7%). The characteristics of patients within each phenotype were significantly different from the original population. The SC was -0.007 and the inclusion of phenotype classification in a regression model did not improve the model performance (0.79 and 0.78 ROC for original and validation model). The post-hoc model performed better than the validation model (SC -0.08). 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.

Association of microbiological factors with mortality in Escherichia coli bacteraemia presenting with sepsis/septic shock: a prospective cohort study.

OBJECTIVES: This study aimed to determine the association of Escherichia coli microbiological factors with 30-day mortality in patients with bloodstream infection (BSI) presenting with a dysregulated response to infection (i.e. sepsis or septic shock). METHODS: Whole-genome sequencing was performed on 224 E coli isolates of patients with sepsis/septic shock, from 22 Spanish hospitals. Phylogroup, sequence type, virulence, antibiotic resistance, and pathogenicity islands were assessed. A multivariable model for 30-day mortality including clinical and epidemiological variables was built, to which microbiological variables were hierarchically added. The predictive capacity of the models was estimated by the area under the receiver operating characteristic curve (AUROC) with 95% confidence intervals (CI). RESULTS: Mortality at day 30 was 31% (69 patients). The clinical model for mortality included (adjusted OR; 95% CI) age (1.04; 1.02-1.07), Charlson index ≥3 (1.78; 0.95-3.32), urinary BSI source (0.30; 0.16-0.57), and active empirical treatment (0.36; 0.11-1.14) with an AUROC of 0.73 (95% CI, 0.67-0.80). Addition of microbiological factors selected clone ST95 (3.64; 0.94-14.04), eilA gene (2.62; 1.14-6.02), and astA gene (2.39; 0.87-6.59) as associated with mortality, with an AUROC of 0.76 (0.69-0.82). DISCUSSION: Despite having a modest overall contribution, some microbiological factors were associated with increased odds of death and deserve to be studied as potential therapeutic or preventive targets.

Corticosteroid treatment in critically ill patients with severe influenza pneumonia: a propensity score matching study.

PURPOSE: To determine clinical predictors associated with corticosteroid administration and its association with ICU mortality in critically ill patients with severe influenza pneumonia. METHODS: Secondary analysis of a prospective cohort study of critically ill patients with confirmed influenza pneumonia admitted to 148 ICUs in Spain between June 2009 and April 2014. Patients who received corticosteroid treatment for causes other than viral pneumonia (e.g., refractory septic shock and asthma or chronic obstructive pulmonary disease [COPD] exacerbation) were excluded. Patients with corticosteroid therapy were compared with those without corticosteroid therapy. We use a propensity score (PS) matching analysis to reduce confounding factors. The primary outcome was ICU mortality. Cox proportional hazards and competing risks analysis was performed to assess the impact of corticosteroids on ICU mortality. RESULTS: A total of 1846 patients with primary influenza pneumonia were enrolled. Corticosteroids were administered in 604 (32.7%) patients, with methylprednisolone the most frequently used corticosteroid (578/604 [95.7%]). The median daily dose was equivalent to 80 mg of methylprednisolone (IQR 60-120) for a median duration of 7 days (IQR 5-10). Asthma, COPD, hematological disease, and the need for mechanical ventilation were independently associated with corticosteroid use. Crude ICU mortality was higher in patients who received corticosteroids (27.5%) than in patients who did not receive corticosteroids (18.8%, p < 0.001). After PS matching, corticosteroid use was associated with ICU mortality in the Cox (HR = 1.32 [95% CI 1.08-1.60], p < 0.006) and competing risks analysis (SHR = 1.37 [95% CI 1.12-1.68], p = 0.001). CONCLUSION: Administration of corticosteroids in patients with severe influenza pneumonia is associated with increased ICU mortality, and these agents should not be used as co-adjuvant therapy.