Elisabeth Papiol

Background/Objectives: The SARS-CoV-2 and influenza A (H1N1)pdm09 pandemics have resulted in high numbers of ICU admissions, with high mortality. Identifying risk factors for ICU mortality at the time of admission can help optimize clinical decision making. However, the risk factors identified may differ, depending on the type of analysis used. Our aim is to compare the risk factors and performance of a linear model (multivariable logistic regression, GLM) with a non-linear model (random forest, RF) in a large national cohort. Methods: A retrospective analysis was performed on a multicenter database including 8902 critically ill patients with influenza A (H1N1)pdm09 or COVID-19 admitted to 184 Spanish ICUs. Demographic, clinical, laboratory, and microbiological data from the first 24 h were used. Prediction models were built using GLM and RF. The performance of the GLM was evaluated by area under the ROC curve (AUC), precision, sensitivity, and specificity, while the RF by out-of-bag (OOB) error and accuracy. In addition, in the RF, the im-portance of the variables in terms of accuracy reduction (AR) and Gini index reduction (GI) was determined. Results: Overall mortality in the ICU was 25.8%. Model performance was similar, with AUC = 76% for GLM, and AUC = 75.6% for RF. GLM identified 17 independent risk factors, while RF identified 19 for AR and 23 for GI. Thirteen variables were found to be important in both models. Laboratory variables such as procalcitonin, white blood cells, lactate, or D-dimer levels were not significant in GLM but were significant in RF. On the contrary, acute kidney injury and the presence of Acinetobacter spp. were important variables in the GLM but not in the RF. Conclusions: Although the performance of linear and non-linear models was similar, different risk factors were determined, depending on the model used. This alerts clinicians to the limitations and usefulness of studies limited to a single type of model.

Background: During the influenza A(H1N1) and COVID-19 pandemics, empirical antibiotic treatment (EAT) was widely administered to critically ill patients despite low rates of confirmed bacterial co-infection (COI). The clinical benefit of this practice remains uncertain and may contradict antimicrobial stewardship principles. Objective: To evaluate whether EAT at ICU admission reduces ventilator-associated pneumonia (VAP) incidence or ICU mortality in critically ill patients with pandemic viral pneumonia, stratified by presence of COI. Methods: This retrospective analysis combined two national multicentre ICU registries in Spain, including 4197 adult patients requiring invasive mechanical ventilation for influenza A(H1N1) or COVID-19 between 2009 and 2021. Primary outcomes were ICU mortality and VAP incidence. Analyses were stratified by microbiologically confirmed bacterial COI. Propensity score matching, Cox regression, General Linear (GLM), and random forest models were applied. Results: Among patients without COI (n = 3543), EAT was not associated with lower ICU mortality (OR = 1.02, 95%CI 0.81–1.28, p = 0.87) or VAP (OR = 1.02, 95%CI 0.79–1.39, p = 0.89). In patients with confirmed COI (n = 654), appropriate EAT was associated with reduced VAP (17.4% vs. 36.3%, p < 0.001) and ICU mortality (38.4% vs. 49.6%, OR = 1.89, 95%CI 1.13–3.14, p = 0.03) compared to inappropriate EAT. Conclusions: EAT was not associated with a lower incidence of VAP or higher survival rates and could be harmful if administered incorrectly. These findings support a more targeted approach to antibiotic use, guided by microbiology, biomarkers and stewardship principles.

Background: The SARS-CoV-2 and influenza A(H1N1)pdm09 pandemics have resulted in high numbers of ICU admis-sions with high mortality. Identifying risk factors for ICU mortality at the time of admission can help opti-mize clinical decision making. However, the risk factors identified may differ depending on the type of analysis used. Our aim is to compare the risk factors and performance of a linear model (multivariable lo-gistic regression, GLM) with a non-linear model (random forest, RF) in a large national cohort. Methods: A retrospective analysis was performed on a multicenter database including 8902 critical patients with influenza A(H1N1)pdm09 or COVID-19 admitted to 184 Spanish ICUs. Demographic, clinical, labora-tory and microbiological data from the first 24 hours were used. Prediction models were built using GLM and RF. The performance of the GLM was evaluated by area under the ROC curve (AUC), precision, sensi-tivity and specificity, while the RF by out-of-bag (OOB) error and accuracy. In addition, in the RF, the im-portance of the variables in terms of accuracy reduction (AR) and Gini index reduction (GI) was deter-mined. Results: Overall mortality in the ICU was 25.8%. Model performance was similar, AUC =76% for GLM and AUC 75.6% for RF. GLM identified 17 independent risk factors, while RF identified 19 for AR and 23 for GI. Thirteen variables were found to be important in both models. Laboratory variables such as procalcitonin, white blood cells, lactate or D-dimer were not significant in GLM but were significant in RF. On the contra-ry, acute kidney injury and the presence of Acinetobacter spp. were important variables in the GLM but not in the RF. Conclusions: Although the performance of linear and non-linear models was similar, different risk factors were determined depending on the model used. This alerts clinicians to the limitations and usefulness of studies limited to a single type of model.