Volatile compounds in human breath: critical review and meta-analysis

Area of Science:

• Biochemistry
• Metabolomics
• Analytical Chemistry

Background:

• Human breath contains volatile compounds reflecting physiological states.
• Current breath biomarker research faces challenges due to inconsistent identification and conflicting results.
• Novel strategies are needed to identify reliable breath biomarkers for disease diagnostics.

Purpose of the Study:

• To group volatile organic compounds (VOCs) from literature data into functional categories based on metabolic pathways.
• To enhance biomarker discovery by analyzing VOCs through a functional grouping approach.
• To improve the accuracy and success rate of diagnostic applications using breath biomarkers.

Main Methods:

• Literature review of VOCs associated with four distinct diseases.
• Grouping identified VOCs into chemical functional groups (e.g., aldehydes, hydrocarbons).
• Application of principal component analysis (PCA), random forest, and linear discriminant analysis (LDA).

Main Results:

• Functional grouping of VOCs doubled explanatory capacity in PCA from 19.1% to 38%.
• Random forest and LDA achieved 93% classification accuracy for cancer detection.
• Identified specific volatile functional groups associated with disease states.

Conclusions:

• A functional grouping approach for VOCs offers superior insight compared to individual compound analysis.
• Targeted, functional volatile biomarkers show promise for improving diagnostic accuracy in research and clinical applications.
• This meta-analysis provides a framework for future research designs in breath analysis for disease detection.