Howard D. Sesso

Background: Anthocyanins and anthocyanin‐rich berries have demonstrated lipid‐lowering and anti‐inflammatory effects; however, associations with cardiovascular disease remain mixed. We therefore examined anthocyanin and blueberry intake for prospective associations with cardiovascular disease risk and cross‐sectional associations with lipids and hs‐CRP (high‐sensitivity C‐reactive protein) in the WHS (Women's Health Study). Methods: Among 38 088 women who completed a baseline semiquantitative food frequency questionnaire in 1992 to 1995, we estimated intake of total anthocyanins, 6 anthocyanin subclasses, and blueberries. Multivariable hazard ratios (HRs) were calculated for confirmed major adverse cardiovascular events, including nonfatal myocardial infarction, nonfatal stroke, and cardiovascular disease death. In a subset of 26 912 women with baseline blood samples, multivariable odds ratios were calculated for adverse lipid and hs‐CRP levels. Results: During a median 28.0‐year follow‐up, 2752 major adverse cardiovascular event cases occurred. Age‐, energy‐, and WHS treatment–adjusted HRs across quintiles of total anthocyanin intake were 1.00 (reference), 0.87 (95% CI, 0.78–0.98), 0.79 (95% CI, 0.70–0.89), 0.81 (95% CI, 0.72–0.91), and 0.88 (95% CI, 0.79–0.99) ( P ‐trend=0.23). Further adjustment for clinical, lifestyle, and dietary factors showed no significant associations with major adverse cardiovascular events or their components, with similar null findings observed for anthocyanin subclasses and blueberries. In cross‐sectional analyses, higher total anthocyanins and anthocyanin subclasses, except pelargonidin, were significantly associated with lower multivariable‐adjusted odds of adverse high‐density lipoprotein cholesterol and triglycerides (all other P ‐trend<0.05). Conclusions: In the WHS, anthocyanin and blueberry intake were not significantly associated with cardiovascular disease risk. However, higher anthocyanin intake was cross‐sectionally associated with lower odds of adverse high‐density lipoprotein cholesterol and triglycerides.

We address concerns regarding inter-individual differences in flavanol metabolism and biomarker levels and explain why they do not affect the conclusions of our study of the impact of background diet and adherence on the outcomes analysis of COSMOS.

Aims: Chronic inflammation associates with prevalent and short-term incident hypertension. The aim of this study is to determine whether baseline high-sensitivity C-reactive protein (hsCRP) independently associates with incident hypertension over three decades in women. Methods and results: Normotensive Women’s Health Study participants with a baseline hsCRP measurement were followed for incident hypertension for up to 30.9 years. Cause-specific Cox proportional hazards models estimated hazard ratios for incident hypertension across hsCRP quintiles and per SD log(hsCRP). Models were adjusted for a minimally sufficient set of covariates derived from a directed acyclic graph, including age, study arm, body mass index (BMI), smoking, exercise, dietary factors, alcohol use, eGFR, and hormone replacement therapy (HRT). Stratified analyses were performed across categories of age, systolic blood pressure (SBP), diastolic blood pressure (DBP), body mass index (BMI), and HRT use. Landmark analyses estimated separate hazard ratios for each decade. 14 234 of 20 224 (70%) women developed hypertension during follow-up. Baseline hsCRP was associated with time-to-incident hypertension, with a time-averaged multivariable hazard ratio of 1.33 (95% CI 1.25–1.41) comparing the top and bottom quintiles, and 1.10 (95% CI 1.08–1.12) per SD log(hsCRP). There was no effect modification by age, SBP, DBP, BMI, or HRT use. In landmark analyses, hsCRP was associated with incident hypertension in the first two decades, but not beyond 20 years. Conclusion: A single measure of low-grade inflammation independently associated with time-to-incident hypertension over the near- to intermediate-term, with limited prognostic value beyond 20 years. These data suggest that inflammation may be a primordial risk factor for hypertension in women.

Background: Obesity increases risks of chronic liver disease and liver cancer, but risks across obesity phenotypes and related molecular signatures remain incompletely characterized. We evaluated associations of obesity phenotypes and obesity-related omics signatures with adverse liver outcomes. Methods: We analyzed 451 463 UK Biobank participants aged 40-69 years. General and central obesity were defined by body mass index and waist circumference. Metabolically unhealthy obesity was defined as obesity plus at least 1 metabolic abnormality. Liver outcomes (metabolic dysfunction–associated steatotic liver disease, cirrhosis, liver cancer, severe liver disease, and chronic liver disease mortality) were ascertained from hospital inpatient records, cancer and death registries. Cox models estimated hazard ratios (HRs) for obesity measures, and internally validated nuclear magnetic resonance metabolomic and Olink proteomic signatures were derived using elastic-net regression. Results: During a median 11-year follow-up, general obesity and central obesity were associated with higher risks of adverse liver outcomes (HR = 1.53-3.29), with generally stronger associations for metabolically unhealthy obesity (HR = 1.69-3.68). Participants with both general and central obesity had the highest risks. Omics signatures of metabolically unhealthy obesity were more strongly associated with liver outcomes than general or central obesity signatures (HR = 1.58-6.13 vs 1.43-4.72). Nuclear magnetic resonance metabolomic biomarkers (albumin, glutamine, triglycerides) and proteomic biomarkers (FABP4, CDHR2, TGFBR2, IL1RN) were consistently associated with multiple liver outcomes. Conclusion: Obesity phenotypes, particularly metabolically unhealthy obesity, and obesity-related omics signatures were associated with higher risks of liver cancer and other adverse liver outcomes. Findings highlight metabolic dysfunction beyond adiposity as an important dimension of obesity-related liver risk and suggest that metabolomic and proteomic profiles may provide molecular insight.

Outcomes from the COSMOS trial have reinforced the notion of flavanols as important plant-derived bioactives contributing to cardiovascular health. As discussions continue on whether specific dietary reference values for flavanols are warranted, it is possible that existing dietary guidelines emphasizing fruits and vegetables already yield sufficient flavanol intake levels. If this were the case, developing flavanol specific dietary reference values might be unnecessary. This study therefore aimed at assessing whether adherence to dietary recommendations for fruit and vegetable intake and overall diet quality achieves flavanol intake levels of 500 mg day⁻¹, the amount proven to mediate cardiovascular benefits in the COSMOS trial. Flavanol intake was objectively evaluated using two validated and complementary biomarkers, 5-(3′,4′-dihydroxyphenyl)-γ-valerolactone metabolites (gVLM_B) and structurally related (−)-epicatechin metabolites (SREM_B), in two geographically distinct studies: COSMOS (US; n = 6509) and EPIC-Norfolk (UK; n = 24 154). The results showed that higher fruit and vegetable intakes and diet quality (assessed via the alternative healthy eating index-aHEI) were associated with increased flavanol intake in COSMOS. Nevertheless, fewer than 25% of participants meeting dietary guidelines achieved an estimated flavanol intake of ≥500 mg day⁻¹. Similar findings were observed in EPIC-Norfolk as well as through flavanol intake simulations considering fruits and vegetables commonly consumed in the US diet. In conclusion, adherence to existing dietary guidelines does not yield flavanol intake levels comparable to those shown to provide cardiovascular benefits in COSMOS. Thus, specific dietary reference values for flavanols may still be necessary if aiming to increase the intake of these dietary compounds.

Despite widespread use in the USA, little is known about the impact of multivitamin-multimineral (MVM) supplementation on metabolomic profiles and how this may translate to biological aging and long-term risk of age-related chronic diseases. The COcoa Supplement and Multivitamin Outcomes Study (COSMOS) is a 2 × 2 factorial, placebo-controlled, double-blinded, randomized clinical trial testing daily MVM (Centrum Silver®) in US older adults. In 399 participants with metabolomics profiles quantified using the Nightingale Health nuclear magnetic resonance platform at baseline, year 1, and year 2 (70.3 ± 5.7 years; 50.1% female), we evaluate the effects of MVM versus placebo on 168 metabolites, 7 metabolomics clocks of biological aging, and 24 metabolomic risk scores (MRS) for chronic diseases. Compared with placebo, 2-year daily MVM supplementation increased the concentrations of docosahexaenoic acid (Cohen's d for between-group differences in changes from baseline to 2 years, 0.212 [95% CI, 0.02 to 0.404]) and omega-3 fatty acids (0.193 [0.003 to 0.382]) as well as delayed the increase in concentrations of creatinine (-0.222 [-0.424 to -0.019]). MVM also reduced MRS for 16 chronic diseases, including major cardiovascular events, renal disease, all-cause dementia, asthma, cataracts, glaucoma, liver disease, non-melanoma skin cancer, and prostate cancer (Cohen's d range, -0.206 to -0.277). However, all these effects were only significant at nominal levels, not after multiple-testing correction (p < 0.05; FDR > 0.05). Additionally, MVM led to a non-significant decrease across all metabolomic aging clocks, ranging from 0.496 to 1.054 years. Despite the limited sample size, COSMOS provides preliminary evidence that daily MVM supplementation may modestly improve metabolomic profiles in older adults. Further expansion within COSMOS is still warranted to confirm whether these alterations contribute to the benefits of MVM supplementation for healthy aging. Trial registration number: NCT02422745.

Cocoa flavanols may reduce cardiovascular disease (CVD) risk, yet large randomized trials remain inconclusive. The COcoa Supplement and Multivitamin Outcomes Study (COSMOS) suggested a modest, nonsignificant benefit using Cox models, which do not account for event severity in composite outcomes. To address this, we applied generalized pairwise comparison (GPC), or “win ratio” (WR), to assess cocoa flavanols versus placebo on hierarchical CVD outcomes among healthy older US adults. This secondary analysis of COSMOS, a randomized, placebo-controlled, 2 × 2 factorial trial of cocoa extract and multivitamins for preventing CVD and cancer, included 21,442 adults (women ≥ 65, men ≥ 60 years) followed for a median of 3.6 years. The primary outcome was a hierarchical composite of total CVD, prioritizing: fatal CVD, non-fatal myocardial infarction (MI), non-fatal stroke, coronary revascularization, carotid surgery, peripheral artery surgery, and hospitalized unstable angina. Analyses followed the intention-to-treat principle. GPC estimated WRs and net treatment benefits (NTBs) for cocoa flavanols versus placebo. GPC analyses showed cocoa flavanol wins of 3.41% and placebo wins of 2.87%, yielding a reciprocal WR of 0.84 (95% CI 0.72–0.99) and negative NTBs of − 0.54% (− 1.04 to − 0.03), p = 0.037. Sensitivity analyses prioritizing stroke over MI produced similar findings. By contrast, Cox regression of the same composite yielded a nonsignificant hazard ratio of 0.90 (95% CI 0.79–1.03), suggesting standard time-to-first-event models underestimated benefit. GPC “WR” analyses showed cocoa flavanols significantly reduced CVD events by accounting for event severity in the composite CVD outcome, whereas Cox regression marginally missed these effects.