Search

Objective: To identify metabolomic biomarkers predictive of Intensive Care Unit (ICU) mortality in adults. Rationale: Comprehensive metabolomic profiling of plasma at ICU admission to identify biomarkers associated with mortality has recently become feasible. Methods: We performed metabolomic profiling of plasma from 90 ICU subjects enrolled in the BWH Registry of Critical Illness (RoCI). We tested individual metabolites and a Bayesian Network of metabolites for association with 28-day mortality, using logistic regression in R, and the CGBayesNets Package in MATLAB. Both individual metabolites and the network were tested for replication in an independent cohort of 149 adults enrolled in the Community Acquired Pneumonia and Sepsis Outcome Diagnostics (CAPSOD) study. Results: We tested variable metabolites for association with 28-day mortality. In RoCI, nearly one third of metabolites differed among ICU survivors versus those who died by day 28 (N = 57 metabolites, p<.05). Associations with 28-day mortality replicated for 31 of these metabolites (with p<.05) in the CAPSOD population. Replicating metabolites included lipids (N = 14), amino acids or amino acid breakdown products (N = 12), carbohydrates (N = 1), nucleotides (N = 3), and 1 peptide. Among 31 replicated metabolites, 25 were higher in subjects who progressed to die; all 6 metabolites that are lower in those who die are lipids. We used Bayesian modeling to form a metabolomic network of 7 metabolites associated with death (gamma-glutamylphenylalanine, gamma-glutamyltyrosine, 1-arachidonoylGPC(20:4), taurochenodeoxycholate, 3-(4-hydroxyphenyl) lactate, sucrose, kynurenine). This network achieved a 91% AUC predicting 28-day mortality in RoCI, and 74% of the AUC in CAPSOD (p<.001 in both populations). Conclusion: Both individual metabolites and a metabolomic network were associated with 28-day mortality in two independent cohorts. Metabolomic profiling represents a valuable new approach for identifying novel biomarkers in critically ill patients. [ABSTRACT FROM AUTHOR]

Background: Metabolic predictors and potential mediators of survival in sepsis have been incompletely characterized. We examined whether machine learning (ML) tools applied to the human plasma metabolome could consistently identify and prioritize metabolites implicated in sepsis survivorship, and whether these methods improved upon conventional statistical approaches. Methods: Plasma gas chromatography–liquid chromatography mass spectrometry quantified 411 metabolites measured ≤ 72 h of ICU admission in 60 patients with sepsis at a single center (Brigham and Women's Hospital, Boston, USA). Seven ML approaches were trained to differentiate survivors from non-survivors. Model performance predicting 28 day mortality was assessed through internal cross-validation, and innate top-feature (metabolite) selection and rankings were compared across the 7 ML approaches and with conventional statistical methods (logistic regression). Metabolites were consensus ranked by a summary, ensemble ML ranking procedure weighing their contribution to mortality risk prediction across multiple ML models. Results: Median (IQR) patient age was 58 (47, 62) years, 45% were women, and median (IQR) SOFA score was 9 (6, 12). Mortality at 28 days was 42%. The models' specificity ranged from 0.619 to 0.821. Partial least squares regression-discriminant analysis and nearest shrunken centroids prioritized the greatest number of metabolites identified by at least one other method. Penalized logistic regression demonstrated top-feature results that were consistent with many ML methods. Across the plasma metabolome, the 13 metabolites with the strongest linkage to mortality defined through an ensemble ML importance score included lactate, bilirubin, kynurenine, glycochenodeoxycholate, phenylalanine, and others. Four of these top 13 metabolites (3-hydroxyisobutyrate, indoleacetate, fucose, and glycolithocholate sulfate) have not been previously associated with sepsis survival. Many of the prioritized metabolites are constituents of the tryptophan, pyruvate, phenylalanine, pentose phosphate, and bile acid pathways. Conclusions: We identified metabolites linked with sepsis survival, some confirming prior observations, and others representing new associations. The application of ensemble ML feature-ranking tools to metabolomic data may represent a promising statistical platform to support biologic target discovery. [ABSTRACT FROM AUTHOR]

Background: Metabolic predictors and potential mediators of survival in sepsis have been incompletely characterized. We examined whether machine learning (ML) tools applied to the human plasma metabolome could consistently identify and prioritize metabolites implicated in sepsis survivorship, and whether these methods improved upon conventional statistical approaches. Methods: Plasma gas chromatography–liquid chromatography mass spectrometry quantified 411 metabolites measured ≤ 72 h of ICU admission in 60 patients with sepsis at a single center (Brigham and Women's Hospital, Boston, USA). Seven ML approaches were trained to differentiate survivors from non-survivors. Model performance predicting 28 day mortality was assessed through internal cross-validation, and innate top-feature (metabolite) selection and rankings were compared across the 7 ML approaches and with conventional statistical methods (logistic regression). Metabolites were consensus ranked by a summary, ensemble ML ranking procedure weighing their contribution to mortality risk prediction across multiple ML models. Results: Median (IQR) patient age was 58 (47, 62) years, 45% were women, and median (IQR) SOFA score was 9 (6, 12). Mortality at 28 days was 42%. The models' specificity ranged from 0.619 to 0.821. Partial least squares regression-discriminant analysis and nearest shrunken centroids prioritized the greatest number of metabolites identified by at least one other method. Penalized logistic regression demonstrated top-feature results that were consistent with many ML methods. Across the plasma metabolome, the 13 metabolites with the strongest linkage to mortality defined through an ensemble ML importance score included lactate, bilirubin, kynurenine, glycochenodeoxycholate, phenylalanine, and others. Four of these top 13 metabolites (3-hydroxyisobutyrate, indoleacetate, fucose, and glycolithocholate sulfate) have not been previously associated with sepsis survival. Many of the prioritized metabolites are constituents of the tryptophan, pyruvate, phenylalanine, pentose phosphate, and bile acid pathways. Conclusions: We identified metabolites linked with sepsis survival, some confirming prior observations, and others representing new associations. The application of ensemble ML feature-ranking tools to metabolomic data may represent a promising statistical platform to support biologic target discovery. [ABSTRACT FROM AUTHOR]