Premium
Principal Components Analysis of Urine Caffeine and Caffeine Metabolites in the US Population
Author(s) -
Pao ChingI,
Rybak Michael,
Sternberg Maya,
Ahluwalia Namanjeet,
Pfeiffer Christine
Publication year - 2015
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.29.1_supplement.586.7
Subject(s) - theobromine , paraxanthine , caffeine , theophylline , chemistry , metabolite , population , urine , principal component analysis , medicine , metabolism , biochemistry , mathematics , cyp1a2 , statistics , environmental health , cytochrome p450
The NHANES 2009–2010 provides the first nationally representative data on urine caffeine and 14 caffeine metabolite concentrations in the US population 蠅6 y (n=2714). We studied the correlation of these compounds with 24‐h caffeine intake (diet and supplements) and observed two apparently distinct groups. “Group 1” (Spearman |r| with intake: 0.55–0.68, P <0.0001), consisted of caffeine, 3 of its metabolites (paraxanthine, theophylline, 1,3,7‐trimethyluric acid) and 5 subsequent metabolites of paraxanthine and/or theophylline. “Group 2” (Spearman |r|: 0.15–0.33, P <0.0001) consisted of theobromine, 3,7‐dimethyluric acid, and related metabolites. We used principal components analysis (PCA) to further describe potential relationships between these apparent groups. After log transformation and centering the data we found that the first two principal components explained 90% of the variability in the entire system. The first principal component (PC1), which can be interpreted as a weighted average of all 15 analytes, explained 77% of the variability and had positive coefficients for all compounds (0.11–0.35, median 0.26). The second principal component (PC2), which explained ~13% of the variability, was a difference between the weighted averages of Group 2 (0.21–0.38, median 0.32) and Group 1 (‐0.34 to‐0.03, median ‐0.19). When caffeine and theobromine intakes were included in the PCA, caffeine intake was associated with Group 1 compounds and theobromine intake with Group 2. These findings support the potential use of the Group 1 compounds as biomarkers of caffeine intake.