Kari Kuulasmaa

BACKGROUND: Stratification of stroke risk remains challenging, but metabolomic profiling offers the potential to detect new biomarkers and to improve early risk assessment of incident stroke. The objective of this study was to evaluate the association between circulating metabolites and the incidence of stroke in a case-cohort study conducted across several large population-based European cohorts. METHODS: Following the case-cohort design, a subset of 10 299 individuals, including all individuals with incident stroke, was selected from the original cohort of >70 000 individuals. The case-cohort design used a random subsample of the selected population cohorts, supplemented with cases not sampled in this random subcohort. A total of 141 circulating metabolites were measured from serum samples of the selected individuals, and associations of these metabolites with risk of incident stroke were estimated and compared with those of classic risk factors (sex, age at examination, systolic blood pressure, total cholesterol, body mass index, diabetes, daily smoking status, and antihypertensive treatment). Associations with time to stroke were assessed using weighted Cox proportional hazards models adjusted for the classic risk factors. Hazard ratios (HRs) for the log-transformed metabolites were reported per 1 SD increase. RESULTS: Of the 70 195 individuals in the original cohort, 1516 (2.2%) experienced incident strokes during a median follow-up time of 8.9 years (interquartile range, 4.4–14.7). Median age was 56.8 years (interquartile range, 49.5–62.4), and 39.5% were women. Six of the 141 metabolites (2 diacyl-phosphatidylcholines, 2 lyso-phosphatidylcholines, 1 hydroxysphingomyelin, and glutamic acid) remained significantly associated with incident stroke after correction for multiple comparisons (adjusted HRs [95% CIs] per SD: lyso-phosphatidylcholines a C18:2 HR, 0.88 [95% CI, 0.82–0.93], lyso-phosphatidylcholines a C17:0 HR, 0.88 [95% CI, 0.83–0.94], hydroxysphingomyelin C14:1 HR, 0.90 [95% CI, 0.85–0.94], diacyl-phosphatidylcholines C34:1 HR, 1.10 [95% CI 1.05–1.16], diacyl-phosphatidylcholines C32:1 HR, 1.14 [95% CI, 1.08–1.21], glutamic acid HR, 1.23 [95% CI 1.11–1.37]). The strengths of these associations were similar to those for classic cardiovascular risk factors (C statistics for 10-year prediction ranging from 0.782 to 0.785 for metabolites compared with 0.781 to 0.792 for classic cardiovascular risk factors). CONCLUSIONS: Among 10 299 individuals from the general European population, we identified 6 metabolites from 4 different metabolite classes that were associated with future risk of stroke. The application of specific circulating metabolites may improve early stroke risk prediction before the onset of potentially irreversible cerebrovascular pathological processes.

Aims: Deleterious effects of Lipoprotein (a) (Lp(a)) might be mitigated by overall cardiovascular (CV) risk reduction. However, data on the relationship between increased Lp(a) and incident coronary heart disease (CHD) according to the distribution of modifiable CV risk factors (CVRF) at baseline are still scarce. We investigated the association between high Lp(a) and incident CHD in the general population, depending on the presence/absence of four major CVRFs (hypertension, diabetes, hypercholesterolemia, smoking) at baseline. Methods and results: Overall 66 495 CHD-free individuals from eight European prospective population-based cohorts were included. The cohort was stratified according to CVRF burden at baseline in ‘0/1 CVRF’ (low risk; n = 41 770) and ‘≥2 CVRFs’ (increased risk; n = 24 725). Fine and Gray competing risk-adjusted models were calculated for the association between Lp(a) mass (<90th vs. ≥90th percentile (pctl.); cut-off 43.2 mg/dL) and future CHD events. During a median follow-up of 9.7 years, 3467 incident CHD events occurred. Despite being at very low absolute risk based on traditional CVRF, individuals with 0/1CVRF demonstrated a strong association between increased Lp(a) mass (≥90th pctl.) and future CHD events, which was comparable to the association observed among individuals with ≥2 CVRFs. The fully-adjusted sub-distribution Hazard Ratios [sHRs] for elevated Lp(a) were 1.38 (95% CI, 1.12–1.71) vs. 1.27 (95% CI, 1.10–1.46) in those having 0/1 vs. ≥2 CVRFs at baseline (Pinteraction 0.50). Conclusion: Among CHD-free subjects, high Lp(a) was related to adverse outcome even in individuals with no or only one CVRF at baseline, thereby generating substantial challenges in mitigating Lp(a)-associated CHD risk in very low risk populations.

Background: It has been hypothesized but seldom tested that the winter excess in cardiovascular disease (CVD) is related to hypovitaminosis D. The present study examined the association between CVD and (i) seasonality of 25-hydroxyvitamin D (25[OH]D) and (ii) individual 25(OH)D concentrations. Methods and findings: Harmonized 25(OH)D data were obtained from the Biomarkers for Cardiovascular Risk Assessment in Europe (BiomarCaRE) project, including 79,570 participants examined between 1984 and 2010. One 25(OH)D measurement was available per participant. Primary endpoints were CVD incidence (coronary heart disease or stroke; n = 6006) and CVD mortality (n = 2985). To study (i), Poisson regression-derived rate ratios were compared according to two-month categories, ordered by baseline 25(OH)D concentrations. To study (ii), Cox regression-derived hazard ratios were compared according to quarters of baseline 25(OH)D concentrations. With respect to (i), despite a median 25(OH)D concentration ratio of 1:1.79, the trough months of 25(OH)D in March and April had a similar CVD incidence as the peak months of 25(OH)D in August and September (rate ratio: 1.07, 95% CI: 0.98–1.17). CVD mortality was slightly higher in the trough months compared to the peak months (rate ratio: 1.27, 95% CI: 1.12–1.44) but not compared to the other months (despite median 25[OH]D concentration ratios up to 1:1.62; p ≥ 0.077). The CVD mortality peak in January preceded the 25(OH)D trough, not adhering to the temporality criterion of Bradford Hill. With respect to (ii), compared to the lowest quarter, the highest quarter of 25(OH)D was associated with lower CVD incidence (hazard ratio: 0.82, 95% CI: 0.76–0.89) and CVD mortality (hazard ratio: 0.64, 95% CI: 0.57–0.72). Conclusion: The present study does not support the hypothesis that seasonal increases in CVD are driven by short-term reductions in 25(OH)D. As in most observational studies, higher 25(OH)D concentrations were inversely associated with CVD.

Introduction: Risk stratification scores such as the European Systematic COronary Risk Evaluation (SCORE) are used to guide individuals on cardiovascular disease (CVD) prevention. Adding high-sensitivity troponin I (hsTnI) to such risk scores has the potential to improve accuracy of CVD prediction. We investigated how applying hsTnI in addition to SCORE may impact management, outcome, and cost-effectiveness. Methods: Characteristics of 72,190 apparently healthy individuals from the Biomarker for Cardiovascular Risk Assessment in Europe (BiomarCaRE) project were included into a discrete-event simulation comparing two strategies for assessing CVD risk. The standard strategy reflecting current practice employed SCORE (SCORE); the alternative strategy involved adding hsTnI information for further stratifying SCORE risk categories (S-SCORE). Individuals were followed over ten years from baseline examination to CVD event, death or end of follow-up. The model tracked the occurrence of events and calculated direct costs of screening, prevention, and treatment from a European health system perspective. Cost-effectiveness was expressed as incremental cost-effectiveness ratio (ICER) in € per quality-adjusted life year (QALYs) gained during 10 years of follow-up. Outputs were validated against observed rates, and results were tested in deterministic and probabilistic sensitivity analyses. Results: S-SCORE yielded a change in management for 10.0% of individuals, and a reduction in CVD events (4.85% vs. 5.38%, p<0.001) and mortality (6.80% vs. 7.04%, p<0.001). S-SCORE led to 23 (95%CI: 20–26) additional event-free years and 7 (95%CI: 5–9) additional QALYs per 1,000 subjects screened, and resulted in a relative risk reduction for CVD of 9.9% (95%CI: 7.3–13.5%) with a number needed to screen to prevent one event of 183 (95%CI: 172 to 203). S-SCORE increased costs per subject by 187€ (95%CI: 177 € to 196 €), leading to an ICER of 27,440€/QALY gained. Sensitivity analysis was performed with eligibility for treatment being the most sensitive. Conclusion: Adding a person’s hsTnI value to SCORE can impact clinical decision making and eventually improves QALYs and is cost-effective compared to CVD prevention strategies using SCORE alone. Stratifying SCORE risk classes for hsTnI would likely offer cost-effective alternatives, particularly when targeting higher risk groups.

Recent investigations have suggested an interdependence of lipoprotein(a) [Lp(a)]-related risk for cardiovascular disease with background inflammatory burden. The aim the present analysis was to investigate whether high-sensitive C-reactive protein (hsCRP) modulates the association between Lp(a) and coronary heart disease (CHD) in the general population. Data from 71 678 participants from 8 European prospective population-based cohort studies were used (65 661 without/6017 with established CHD at baseline; median follow-up 9.8/13.8 years, respectively). Fine and Gray competing risk-adjusted models were calculated according to accompanying hsCRP concentration (P_interaction = 0.82). In those with established CHD, similar associations were seen only among individuals with hsCRP ≥ 2 mg/L [1.34 (1.03–1.76)], whereas among participants with a hsCRP concentration P_interaction = 0.024). While among CHD-free individuals Lp(a) was significantly associated with incident CHD regardless of hsCRP, in participants with CHD at baseline, Lp(a) was related to recurrent CHD events only in those with residual inflammatory risk. These findings might guide adequate selection of high-risk patients for forthcoming Lp(a)-targeting compounds. C-reactive protein as a possible modifier of lipoprotein(a)-related risk for coronary heart disease in Europe. CHD, coronary heart disease; CI, confidence interval; Lp(a), lipoprotein(a); FU, follow-up; F, fifth; hsCRP, high-sensitive C-reactive protein; sHR, sub-distribution hazard ratio.

Data from a single study are rarely sufficient to comprehensively understand a phenomenon of interest or to build predictive models with statistical power. Meta-analyses, including aggregated data for several cohort studies, can be carried out to synthesize published results, but the use of harmonized individual-level data from different studies is preferable to reach the maximal comparability across the studies. A significant barrier to the co-analysis approach is finding relevant studies and accessing their data. Such barriers have, amongst others, led to designing the FAIR (Findable, Accessible, Interoperable and Reusable) principles, which have been recognized internationally. ¹ Currently, it is still difficult to obtain a comprehensive overview of available data from existing observational studies on cardiovascular diseases. Because there is no requirement to register studies other than clinical trials, information on existing studies and their characteristics is often lacking or highly fragmented. In addition, it is often unclear how to request access to the raw data or what kind of ethical, legal and societal issues (ELSI) may be related to the use of the data. All these factors are substantial hindrances to research in the cardiovascular domain and beyond.

Aims: The regional and temporal differences in the associations between cardiovascular disease (CVD) and its classic risk factors are unknown. The current study examined these associations in different European regions over a 30-year period. Methods and results: The study sample comprised 553 818 individuals from 49 cohorts in 11 European countries (baseline: 1982–2012) who were followed up for a maximum of 10 years. Risk factors [sex, smoking, diabetes, non-HDL cholesterol, systolic blood pressure (BP), and body mass index (BMI)] and CVD events (coronary heart disease or stroke) were harmonized across cohorts. Risk factor–outcome associations were analysed using multivariable-adjusted Cox regression models, and differences in associations were assessed using meta-regression. The differences in the risk factor–CVD associations between central Europe, northern Europe, southern Europe, and the UK were generally small. Men had a slightly higher hazard ratio (HR) in southern Europe (P = 0.043 for overall difference), and those with diabetes had a slightly lower HR in central Europe (P = 0.022 for overall difference) compared with the other regions. Of the six CVD risk factors, minor HR decreases per decade were observed for non-HDL cholesterol [7% per mmol/L; 95% confidence interval (CI), 3–10%] and systolic BP (4% per 20 mmHg; 95% CI, 1–8%), while a minor HR increase per decade was observed for BMI (7% per 10 kg/m2; 95% CI, 1–13%). Conclusion: The results demonstrate that all classic CVD risk factors are still relevant in Europe, irrespective of regional area. Preventive strategies should focus on risk factors with the greatest population attributable risk.

Aims: The role of biomarkers in predicting cardiovascular outcomes in high-risk individuals is not well established. We aimed to investigate benefits of adding biomarkers to cardiovascular risk assessment in individuals with and without diabetes. Methods and results: We used individual-level data of 95 292 individuals of the European population harmonized in the Biomarker for Cardiovascular Risk Assessment across Europe consortium and investigated the prognostic ability of high-sensitivity cardiac troponin I (hs-cTnI), N-terminal prohormone of brain natriuretic peptide (NT-proBNP), and high-sensitivity C-reactive protein (hs-CRP). Cox-regression models were used to determine adjusted hazard ratios of diabetes and log-transformed biomarkers for fatal and non-fatal cardiovascular events. Models were compared using the likelihood ratio test. Stratification by specific biomarker cut-offs was performed for crude time-to-event analysis using Kaplan–Meier plots. Overall, 6090 (6.4%) individuals had diabetes at baseline, median follow-up was 9.9 years. Adjusting for classical risk factors and biomarkers, diabetes [HR 2.11 (95% CI 1.92, 2.32)], and all biomarkers (HR per interquartile range hs-cTnI 1.08 [95% CI 1.04, 1.12]; NT-proBNP 1.44 [95% CI 1.37, 1.53]; hs-CRP 1.27 [95% CI 1.21, 1.33]) were independently associated with cardiovascular events. Specific cut-offs for each biomarker identified a high-risk group of individuals with diabetes losing a median of 15.5 years of life compared to diabetics without elevated biomarkers. Addition of biomarkers to the Cox-model significantly improved the prediction of outcomes (likelihood ratio test for nested models P < 0.001), accompanied by an increase in the c-index (increase to 0.81). Conclusion: Biomarkers improve cardiovascular risk prediction in individuals with and without diabetes and facilitate the identification of individuals with diabetes at highest risk for cardiovascular events.