Early Mortality Prediction in Intensive Care Unit Patients Based on Serum Metabolomic Fingerprint
2024

Predicting Early Mortality in ICU Patients Using Serum Metabolomic Fingerprints

Sample size: 44 publication 10 minutes Evidence: moderate

Author Information

Author(s): Rúben Araújo, Luís Ramalhete, Cristiana P. Von Rekowski, Tiago A. H. Fonseca, Luís Bento, Cecília R. C. Calado, Hartmut Schlüter

Primary Institution: NOVA Medical School, Universidade NOVA de Lisboa

Hypothesis

Can serum metabolomic fingerprints obtained through FTIR spectroscopy improve mortality prediction models for COVID-19 ICU patients?

Conclusion

The study found that FTIR spectroscopy can effectively identify metabolic changes associated with mortality in ICU patients, suggesting its potential as a diagnostic tool.

Supporting Evidence

  • FTIR spectroscopy identified significant spectral differences between discharged and deceased patients.
  • Naïve Bayes models achieved AUCs of 0.79, 0.97, and 0.98 for mortality prediction at different time points.
  • Combining multiple biomarkers improved predictive accuracy in ICU settings.

Takeaway

Doctors can use a special test to look at blood samples from COVID-19 patients in the ICU to help predict who might get better and who might not.

Methodology

The study analyzed serum samples from ICU patients using FTIR spectroscopy and machine learning models to predict mortality.

Potential Biases

Potential biases may arise from the small sample size and the specific patient population studied.

Limitations

The study's findings need further validation in larger, more diverse cohorts.

Participant Demographics

The study included 44 ICU patients, with 23 deceased and 21 discharged, comprising 17 females and 27 males.

Statistical Information

P-Value

p<0.01

Statistical Significance

p<0.01

Digital Object Identifier (DOI)

10.3390/ijms252413609

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication