Table 3.
Association between coffee-related metabolites and risk of type 2 diabetes in the NHS and NHSII*
| HMDB ID | Metabolite | Super class | 207 case vs. 621 control subjects | ||
|---|---|---|---|---|---|
| OR per 1 SD (95% CI) | P | FDR | |||
| Positively associated with coffee | |||||
| HMDB00610 | C18:2 CE | Lipids and lipid-like molecules | 0.59 (0.48, 0.72) | <0.0001 | <0.001 |
| HMDB00918 | C18:1 CE | Lipids and lipid-like molecules | 0.60 (0.50, 0.73) | <0.0001 | <0.001 |
| HMDB06726 | C20:4 CE | Lipids and lipid-like molecules | 0.61 (0.50, 0.75) | <0.0001 | <0.001 |
| HMDB00875 | Trigonelline | Alkaloids and derivatives | 0.87 (0.70, 1.08) | 0.22 | 0.28 |
| HMDB00630 | Cytosine | Organoheterocyclic compounds | 0.93 (0.77, 1.13) | 0.46 | 0.49 |
| HMDB04400 | AAMU | Organic nitrogen compounds | 1.04 (0.82, 1.33) | 0.74 | 0.76 |
| HMDB01991 | 7-methylxanthine | Organoheterocyclic compounds | 1.09 (0.87, 1.35) | 0.46 | 0.49 |
| HMDB13678 | 4-hydroxyhippuric acid | Benzenoids | 1.12 (0.92, 1.37) | 0.26 | 0.33 |
| HMDB01847 | Caffeine | Organoheterocyclic compounds | 1.18 (0.93, 1.49) | 0.17 | 0.27 |
| HMDB11103 | 1,7-dimethyluric acid | Organoheterocyclic compounds | 1.22 (0.96, 1.56) | 0.11 | 0.19 |
| HMDB00062 | l-carnitine | Organic nitrogen compounds | 1.25 (0.99, 1.57) | 0.06 | 0.12 |
| Negatively associated with coffee | |||||
| HMDB05377 | C50:2 TAG | Lipids and lipid-like molecules | 2.06 (1.60, 2.65) | <0.0001 | <0.001 |
| HMDB07102 | C34:1 DAG | Lipids and lipid-like molecules | 2.03 (1.60, 2.58) | <0.0001 | <0.001 |
| HMDB05369 | C52:2 TAG | Lipids and lipid-like molecules | 1.99 (1.52, 2.61) | <0.0001 | <0.001 |
| HMDB07132 | C34:3 DAG | Lipids and lipid-like molecules | 1.83 (1.40, 2.41) | <0.0001 | <0.001 |
| HMDB07218 | C36:2 DAG | Lipids and lipid-like molecules | 1.81 (1.43, 2.29) | <0.0001 | <0.001 |
| HMDB07103 | C34:2 DAG | Lipids and lipid-like molecules | 1.80 (1.43, 2.28) | <0.0001 | <0.001 |
| HMDB05433 | C50:3 TAG | Lipids and lipid-like molecules | 1.72 (1.33, 2.22) | <0.001 | <0.001 |
| HMDB05384 | C52:3 TAG | Lipids and lipid-like molecules | 1.46 (1.14, 1.87) | 0.003 | 0.01 |
| HMDB05435 | C50:4 TAG | Lipids and lipid-like molecules | 1.39 (1.09, 1.76) | 0.01 | 0.02 |
| HMDB10471 | C50:5 TAG | Lipids and lipid-like molecules | 1.38 (1.10, 1.73) | 0.01 | 0.01 |
| HMDB07219 | C36:3 DAG | Lipids and lipid-like molecules | 1.36 (1.10, 1.70) | 0.01 | 0.01 |
| HMDB10497 | C50:6 TAG | Lipids and lipid-like molecules | 1.35 (1.09, 1.67) | 0.01 | 0.01 |
| HMDB10517 | C52:7 TAG | Lipids and lipid-like molecules | 1.17 (0.95, 1.45) | 0.14 | 0.24 |
| HMDB05436 | C52:6 TAG | Lipids and lipid-like molecules | 1.16 (0.93, 1.45) | 0.20 | 0.28 |
| HMDB10518 | C54:8 TAG | Lipids and lipid-like molecules | 0.99 (0.80, 1.22) | 0.91 | 0.91 |
| HMDB10498 | C54:9 TAG | Lipids and lipid-like molecules | 0.90 (0.73, 1.12) | 0.34 | 0.39 |
| HMDB05448 | C56:9 TAG | Lipids and lipid-like molecules | 0.89 (0.71, 1.10) | 0.28 | 0.33 |
| HMDB10513 | C56:10 TAG | Lipids and lipid-like molecules | 0.87 (0.71, 1.08) | 0.20 | 0.28 |
| HMDB11526 | C22:6 LPE | Lipids and lipid-like molecules | 0.87 (0.70, 1.07) | 0.18 | 0.28 |
| PCA (proportion)† | |||||
| First component (44.9%) | 1.15 (1.08, 1.22) | <0.0001 | |||
| Second component (15.1%) | 0.80 (0.71, 0.89) | <0.0001 | |||
| Third component (10.3%) | 0.96 (0.86, 1.08) | 0.49 | |||
Conditional logistic regression analysis was conducted to adjust for age at blood draw (continuous), race (White, non-White), cumulative average levels of BMI (<23.0, 23.0–24.9, 25.0–27.4, 27.5–29.9, ≥30.0 kg/m2), physical activity (<3.0, 3.0–8.9, 9.0–17.9, 18.0–26.9, ≥27.0 MET h/week), alcohol consumption (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, ≥15.0 g/day), smoking status (never smoker; past smoker, <30 pack-years; past smoker, ≥30 pack-years; current smoker, <30 pack-years; current smoker, ≥30 pack-years), whole grains (quintiles), fruits (quintiles), vegetables (quintiles), polyunsaturated fat–to–saturated fat ratio (quintiles), fish (quintiles), red meat (quintiles), sugar-sweetened beverages (quintiles), total energy intake (quintiles), regular multivitamin use (yes, no), regular aspirin/nonsteroidal anti-inflammatory drug (NSAID) use (yes, no), family history of diabetes (yes, no), menopausal status (premenopausal, postmenopausal, unknown), and menopausal hormone therapy (never or past or current use).
PCA was performed for 30 coffee-related metabolites, and the top three components were linked to diabetes risk.