Dataset Information

Predicting lethal courses in critically ill COVID-19 patients using a machine learning model trained on patients with non-COVID-19 viral pneumonia.

ABSTRACT: In a pandemic with a novel disease, disease-specific prognosis models are available only with a delay. To bridge the critical early phase, models built for similar diseases might be applied. To test the accuracy of such a knowledge transfer, we investigated how precise lethal courses in critically ill COVID-19 patients can be predicted by a model trained on critically ill non-COVID-19 viral pneumonia patients. We trained gradient boosted decision tree models on 718 (245 deceased) non-COVID-19 viral pneumonia patients to predict individual ICU mortality and applied it to 1054 (369 deceased) COVID-19 patients. Our model showed a significantly better predictive performance (AUROC 0.86 [95% CI 0.86-0.87]) than the clinical scores APACHE2 (0.63 [95% CI 0.61-0.65]), SAPS2 (0.72 [95% CI 0.71-0.74]) and SOFA (0.76 [95% CI 0.75-0.77]), the COVID-19-specific mortality prediction models of Zhou (0.76 [95% CI 0.73-0.78]) and Wang (laboratory: 0.62 [95% CI 0.59-0.65]; clinical: 0.56 [95% CI 0.55-0.58]) and the 4C COVID-19 Mortality score (0.71 [95% CI 0.70-0.72]). We conclude that lethal courses in critically ill COVID-19 patients can be predicted by a machine learning model trained on non-COVID-19 patients. Our results suggest that in a pandemic with a novel disease, prognosis models built for similar diseases can be applied, even when the diseases differ in time courses and in rates of critical and lethal courses.

SUBMITTER: Lichtner G

PROVIDER: S-EPMC8225662 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Micronutrient deficiencies can develop in critically ill patients, arising from factors such as decreased intake, increased losses, drug interactions, and hypermetabolism. These deficiencies may compromise important immune functions, with potential implications for patient outcomes. Alternatively, micronutrient blood levels may become low due to inflammation-driven redistribution rather than consumption. This explorative pilot study investigates blood micronutrient concentrations during the first three weeks of ICU stay in critically ill COVID-19 patients and evaluates the impact of additional micronutrient administration. Moreover, associations between inflammation, disease severity, and micronutrient status were explored. We measured weekly concentrations of vitamins A, B6, D, and E; iron; zinc; copper; selenium; and CRP as a marker of inflammation state and the SOFA score indicating disease severity in 20 critically ill COVID-19 patients during three weeks of ICU stay. Half of the patients received additional (intravenous) micronutrient administration. Data were analyzed with linear mixed models and Pearson's correlation coefficient. High deficiency rates of vitamins A, B6, and D; zinc; and selenium (50-100%) were found at ICU admission, along with low iron status. After three weeks, vitamins B6 and D deficiencies persisted, and iron status remained low. Plasma levels of vitamins A and E, zinc, and selenium improved. No significant differences in micronutrient levels were found between patient groups. Negative correlations were identified between the CRP level and levels of vitamins A and E, iron, transferrin, zinc, and selenium. SOFA scores negatively correlated with vitamin D and selenium levels. Our findings reveal high micronutrient deficiency rates at ICU admission. Additional micronutrient administration did not enhance levels or expedite their increase. Spontaneous increases in vitamins A and E, zinc, and selenium levels were associated with inflammation resolution, suggesting that observed low levels may be attributed, at least in part, to redistribution rather than true deficiencies.

Dataset Information

Predicting lethal courses in critically ill COVID-19 patients using a machine learning model trained on patients with non-COVID-19 viral pneumonia.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets