Photo Credit: Stockbusters
The following is a summary of “Breath metabolomics for diagnosis of acute respiratory distress syndrome,” published in the March 2024 issue of Critical Care by Zhang et al.
Researchers conducted a retrospective study to assess if breath metabolites could accurately diagnose acute respiratory distress syndrome (ARDS) in ICU patients requiring invasive ventilation.
They conducted a two-center observational study involving critically ill patients on invasive ventilation. Gas chromatography and mass spectrometry (GC-MS) measured exhaled metabolites, while three experts categorized patients based on the Berlin definition of ARDS. The derivation cohort, comprising patients labeled as “certain” from one hospital, was used to train a classifier based on the five most significant breath metabolites. The diagnostic accuracy of this classifier was then evaluated in all patients from the second hospital, and it was combined with the lung injury prediction score (LIPS).
The results showed that 499 patients participated in this study. The derivation cohort comprised 357 patients (including 60 with certain ARDS; 17%), while the validation cohort consisted of 142 patients (including 47 with certain ARDS; 33%). The classifier featured metabolites such as 1-methylpyrrole, 1,3,5-trifluorobenzene, methoxyacetic acid, 2-methylfuran, and 2-methyl-1-propanol. In the derivation group, the classifier attained an AUROCC of 0.71 (95% CI 0.63–0.78), whereas, in the validation set, it scored 0.63 (95% CI 0.52–0.74). However, merging the breath test with the LIPS did not significantly boost diagnostic accuracy.
Investigators concluded that while an exhaled breath test showed some promise for diagnosing ARDS, its accuracy wasn’t high enough for standalone clinical use, even when combined with a scoring system.
Source: ccforum.biomedcentral.com/articles/10.1186/s13054-024-04882-7