ABSTRACT
Objective Following a Mediterranean diet (MedDiet) and being physically active decrease cardiovascular risk. However, their individual and combined effects on atherothrombosis biomarkers are unexplored.
Approach and Results In 358 random volunteers from the PREDIMED trial (Prevención con Dieta Mediterránea), we assessed the 1-year effects of an intervention with MedDiet, enriched with virgin olive oil (MedDiet-VOO; N=120) or nuts (MedDiet-Nuts; N=119) versus a low-fat control diet (N=119), on atherothrombosis markers. In a secondary, observational approach, we studied the associations of 1-year changes in MedDiet adherence, physical activity (PA) levels, and their combination, with 1-year differences in atherothrombosis biomarkers. The MedDiet-VOO intervention increased HDL platelet activating factor-acetylhydrolase activity by 8.2% (95% CI: 0.92; 15.6) and decreased HDL-bound α1-antitrypsin levels by 5.8% (95% CI: −11.6; −0.046) versus control diet. The MedDiet-Nuts intervention reduced non-esterified fatty acid levels by 9.1% (95% CI: −18.0; −0.19) relative to control. Only the low-fat diet was associated with increases in platelet factor-4 and prothrombin factor1+2 concentrations versus baseline (P=0.012 and P=0.003, respectively). Increments in MedDiet adherence (+1 point) were associated with lower fibrinogen levels (−1.8%, 95% CI: −3.38; −0.17). Increases in PA (+100 METs·min/d) were linked to lower prothrombin factor1+2 (−5.2%, 95% CI: −10.4; −0.077) and plasminogen activator inhibitor-1 concentrations (−3.0%, 95% CI: −5.93; −0.12). Finally, 1-point increments in MedDiet adherence plus 100 METs·min/d of PA were related to lower D-dimer levels (−1.52%, 95% CI: −3.10; 0.026; P-interaction=0.047).
Conclusions Following a MedDiet and increasing PA were associated with improvements in atherothrombosis biomarkers in high cardiovascular risk individuals.
HIGHLIGHTS
1 year of intervention with MedDiet enhanced HDL antithrombotic properties and decreased the levels of non-esterified fatty acids in high cardiovascular risk individuals.
The low-fat control diet was the only intervention linked to increments in the levels of platelet factor-4 and prothrombin activation relative to baseline.
Increases in MedDiet adherence were associated with decrements in fibrinogen levels, increments in PA were linked to decreases in prothrombin fragment 1+2 and PAI-1 levels, and combined increases in MedDiet adherence and PA were related to improvements in D-dimer concentrations.
INTRODUCTION
Thrombosis consists of the formation of a clot to avoid blood loss in the case of a vessel disruption. It is based on the occurrence of four phenomena (endothelial lesion, platelet aggregation, coagulation cascade, and fibrinolysis) which, in turn, depend on a chain of intertwined biochemical responses1. However, its excessive activation is linked to higher cardiovascular risk1. Increased levels of biomarkers related to these responses (or low concentrations, in the case of fibrinolysis) are linked to an increased risk of suffering atherosclerotic events. This association has been reported for plasma biomarkers of endothelial damage (von Willebrand factor2), platelet aggregation (P-selectin3, platelet factor-44), activation of coagulation (fibrinogen5, prothrombin fragment 1+2 –proportional to thrombin formation–6), and fibrinolysis (plasminogen activator inhibitor-1 –PAI1–7, D-dimer8). Other substances (such as non-esterified fatty acids, NEFAs) promote the development of cardiovascular diseases through their capacity to induce endothelial stress and platelet activation9. Finally, antithrombotic properties have also been attributed to high-density lipoproteins (HDLs)10. These lipoproteins have been shown to be involved in the regulation of inflammatory responses (directly related to thrombosis5) and carry enzymes able to metabolize pro-thrombotic molecules such as the platelet activating factor (the platelet activating factor acetylhydrolase, PAF-AH)11.
Better conformity with a traditional Mediterranean Diet (MedDiet) and the usual practice of leisure-time physical activity (PA) are both able to prevent major cardiovascular clinical outcomes12,13. These beneficial effects of MedDiet and PA may be mediated by improvements in risk factors related to blood pressure, glucose metabolism, oxidative stress, and low-grade inflammation14,15. Nevertheless, little evidence related to their effects on atherothrombosis mechanisms is available. In the context of the PREDIMED (Prevención con Dieta Mediterránea) trial (intervention study with MedDiet on high cardiovascular risk individuals), following this dietary pattern was associated with a decrease in P-selectin levels16. Beyond this study, only a small-scale prospective analysis with 21 young, healthy male volunteers reported that following a MedDiet-like dietary pattern was related to lower fibrinogen levels and an attenuation of the coagulation response17. In relation to exercise, very few studies have assessed its relationship with atherothrombosis biomarkers18 and no long-term association between PA and these indicators is available in high cardiovascular risk individuals.
Our main aim was to assess whether a 1-year intervention with MedDiet improved a comprehensive set of atherothrombosis biomarkers in high cardiovascular risk individuals. Our secondary aim was to determine whether 1-year changes in MedDiet adherence, PA, and their combination were associated with 1-year differences in atherothrombosis indicators in this population.
MATERIALS AND METHODS
Study design
Study subjects belonged to the PREDIMED Study. It was a large-scale, parallel, multicenter, randomized controlled trial assessing the long-term effects of following supplemented MedDiets on the primary prevention of cardiovascular disease in high cardiovascular risk individuals12. Volunteers presented type 2 diabetes mellitus or at least three of these risk factors: high low-density lipoprotein cholesterol levels, low HDL cholesterol concentrations, hypertension, overweight/obesity, current smoking, or family history of premature heart disease. Further details of inclusion/exclusion criteria are available in Supplemental Methods. The trial protocol was approved by local institutional ethic committees, registered under the International Standard Randomized Controlled Trial Number ISRCTN35739639 (http://www.isrctn.com/ISRCTN35739639), and described in previous publications12. An institutional ethic committee (CEIC-PSMAR) also approved the protocol of the present study. All participants provided written informed consent before joining the study.
Individuals were randomly assigned to one of three interventions (in a 1:1:1 ratio): 1) a MedDiet enriched with virgin olive oil (MedDiet-VOO); 2) a MedDiet enriched with mixed nuts (MedDiet-Nuts); and 3) a low-fat control diet. MedDiet interventions promoted the consumption of plant-based products and fish, the elimination of animal fats and processed foods, the substitution of red/processed meats for poultry, the intake of virgin olive oil as main culinary fat, and the consumption of foods cooked by traditional methods. Further details of the dietary intervention are available in Supplemental Methods.
For the present analyses, we selected a random subsample of 358 subjects (4.8% of the total PREDIMED population) recruited in Hospital Clinic (Barcelona) and Hospital del Mar Medical Research Institute (Barcelona) study sites, with citrate plasma samples collected at baseline and after 1 year of intervention. 120 volunteers were allocated to the MedDiet-VOO intervention group, 119 to MedDiet-Nuts, and 119 to the low-fat control diet. Samples were stored at −80°C until the analyses. The CONSORT checklist regarding our study is available in Supplemental Table 1.
Description of the study population
Secondary exposure assessments and covariates
MedDiet adherence was estimated at baseline and after 1 year of intervention using the MedDiet adherence score. It was a validated short screener interrogating whether the volunteer followed the essential 14 dietary characteristics associated with a MedDiet (positively scoring: the intake of virgin olive oil, nuts, fruits, vegetables, legumes, fish, and wine in moderation; and the substitution or avoidance of animal fats, red and processed meats, processed foods, and sugary drinks)19. 1-year increments in MedDiet adherence were calculated by subtracting the baseline adherence to the 1-year value.
Leisure-time PA was assessed at baseline and after 1 year by the self-administered Minnesota Leisure-Time Physical Activity Questionnaire, validated in Spanish men and women20,21. The questionnaire interrogated the participants about the number of days and minutes/day they performed 67 different activities in the previous year. Leisure-time PA was quantified in metabolic equivalents of task in minutes per day (METs·min/d) by multiplying the metabolic equivalents of task linked to an activity with its mean duration reported in minutes/day. 1-year changes in PA were calculated as described for MedDiet adherence.
Data on age, sex, educational level, body mass index, smoking habit, and presence of type-2 diabetes mellitus, hypercholesterolemia, hypertriglyceridemia, and hypertension were gathered at baseline by trained clinical personnel12.
Atherothrombosis biomarkers
We used ELISA kits to quantify the levels of: fibrinogen (Human Fibrinogen SimpleStep ELISA kit, ref.: ab208036, abcam, UK), PAI-1 (Human PAI1 SimpleStep ELISA kit (SERPINE1), ref.: ab184863, abcam, UK), platelet factor-4 (PF4 (CXCL4) Human SimpleStep ELISA kit, ref.:ab189573, abcam, UK), prothrombin fragment 1+2 (Human F1+2 (prothrombin fragment 1+2) ELISA kit, ref.: E-EL-H1793, Elabscience, USA), and von Willebrand factor (Human von Willebrand factor SimpleStep ELISA kit, ref.: ab223864, abcam, UK). We assessed the concentrations of D-dimer (D-Dimer, ref.: SR-1709231, Spinreact, Spain) by immunoturbidimetry, and of NEFAs (Non-Esterified Fatty Acids, ref.: FA115, Randox, Spain) using a colorimetric technique in an ABX-Pentra 400 autoanalyzer (Horiba-ABX, France). All the previous analyses were performed in citrate plasma samples. In parallel, in apolipoprotein B-depleted plasma samples (specimen in which all lipoproteins but HDLs were eliminated, prepared from citrate plasma as well22), we quantified total cholesterol by colorimetry (Cholesterol Enzymatic-Colorimetry, ref.: SR-41022, Spinreact, Spain) and α1-antitrypsin (a-1-antitrypsin, ref.: SR-1102054, Spinreact, Spain) by immunoturbidimetry in an ABX-Pentra 400 autoanalyzer (Horiba-ABX, France). With this data, we calculated the α1-antitrypsin/cholesterol ratio. We also determined in these samples the activity of the PAF-AH enzyme by a colorimetric kit (PAF Acetylhydrolase Activity Assay Kit, ref.: K765-1000, Cayman Chemical, USA). A detailed explanation of the quality control and inter-assay coefficients of variation is available in Supplemental Methods and Supplemental Table 2.
Sample size
We assumed that the variability of the measurement of P-selectin in a previous PREDIMED sub-analysis would be similar to that in our variables. A sample size of 119 individuals per group (total sample size: N=357) allowed ≥80% power to detect a 10% change relative to the mean baseline value of P-selectin levels (7.65 ng/mL) between post- and pre-intervention values, and a 14% change relative to the mean baseline value (10.8 ng/mL) in the differences among the three diets. We considered a 2-sided type I error of 0.05, a loss rate of 5%, and the standard deviation of the differences in P-selectin values in this intervention (SD=28.9 ng/mL)16.
Statistical analyses
We first checked the distribution of continuous variables in normality plots and by the Shapiro-Wilk test. We assessed whether there were differences in baseline values among the three intervention groups with one-way ANOVA tests for normally distributed continuous variables, Kruskal-Wallis tests for non-normally distributed continuous ones, and chi-squared tests for categorical parameters.
As the main analysis based on randomization, we evaluated whether there were differences in the 1-year changes in atherothrombosis biomarkers (calculated by subtracting baseline from post-intervention values) in the MedDiet interventions relative to the control diet by multivariate linear regressions. We also analyzed if there were differences between pre- and post-intervention values in every study arm by paired t tests in normally distributed variables and Wilcoxon signed-rank tests in non-normally distributed ones.
As secondary, observational analyses, we studied two exposures assessed with repeated measurements at baseline and at 1-year: 1) changes in MedDiet adherence (1-point increments in the MedDiet adherence score); and 2) PA levels (+100 METs·min/d). The outcomes were the concurrent 1-year differences in atherothrombosis biomarkers. We applied multivariate linear regression models to assess these associations with adjustment for potential confounding. To determine the effect of the combination of increasing 1-point in MedDiet adherence together with 100 METs·min/d, we computed the product of the previous two variables and assessed its relationship with 1-year changes in atherothrombosis indicators in additional linear regression analyses. In addition, we used a likelihood ratio test between a model with and without the interaction product-term “1-year change in MedDiet adherence × 1-year change in PA levels” to determine whether their interaction was significant.
To help in data interpretation, we present the results as percentage changes relative to baseline values. To calculate them, we divided the adjusted difference coefficients obtained in the regression models by the baseline value of the variable in each of the groups (in the main analyses) or in the 358 volunteers (in the secondary analyses). Models were adjusted for: age (continuous), sex, study site, educational level (primary education or lower/secondary or equivalent/higher education), baseline value of the respective atherothrombosis biomarker, type-2 diabetes (yes/no), hypercholesterolemia (yes/no), hypertriglyceridemia (yes/no), hypertension (yes/no), tobacco use (never/former/actual smoker), body mass index (continuous), PA (continuous), alcohol use (continuous), energy intake (continuous), and two propensity scores to correct for the theoretical deviations in the randomization process (calculated from 30 baseline variables)12. Secondary, observational analyses were additionally adjusted for the allocated intervention group.
MedDiet adherence analyses were further adjusted for compliance at baseline. Models were plotted using the “lme” package in R Software23.
We performed statistical analyses with R software version 3.5.024.
RESULTS
Study population and dietary intervention
We found no significant differences in clinical variables at baseline among individuals allocated to three intervention groups (Table 1). Regarding the PREDIMED Study subjects not included in this sub-study, there were 6.2% more women and 7.2% more hypercholesterolemic individuals, and they presented higher PA levels (+22 METs·min/d) and alcohol use (+0.67 g/d) (P<0.05) (Supplemental Table 3).
Volunteers’ compliance with the intended dietary interventions was good according to the MedDiet adherence score: we observed a 1.14-(MedDiet-VOO) and 1.35-point (MedDiet-Nuts) increment in score values after 1 year of intervention (P<0.001, versus control diet) (Supplemental Table 4). Volunteers allocated to the MedDiet-VOO intervention increased their consumption of virgin olive oil, fruits and vegetables, legumes, and fish. Those in the MedDiet-Nuts group incremented their intake of nuts, and fatty fish, and decreased their consumption of processed meat (P<0.05, versus control diet).
Main analyses based on randomization: effects of 1-year MedDiet interventions on atherothrombosis biomarkers
The MedDiet-VOO intervention increased the activity of HDL platelet activating factor-acetylhydrolase (+8.3%, 95% CI: 0.92; 15.6 –Figure 1H–) and decreased the levels of α1-antitrypsin in HDL (−5.8%, 95% CI: −11.6; −0.046 –Figure 1I–) compared to the control group. The MedDiet-Nuts intervention diminished the concentrations of NEFAs versus the control diet (−9.1%, 95% CI: −18.0; −0.19 –Figure 1G–). When assessing post-vs. pre-intervention changes, only the low-fat control diet was associated with increases in platelet factor-4 concentrations (P=0.012) and prothrombin fragment 1+2 levels (P=0.003) relative to baseline (Supplemental Table 5). Exact coefficients are available in Supplemental Table 6.
Effects of MedDiet-VOO and MedDiet-Nuts interventions versus the low-fat control diet (first forest plot of each panel), and associations with 1-year changes in MedDiet adherence (+1 point in MedDiet adherence score), PA (+100 METs·min/d), and their combination (second forest plot): von Willebrand factor (A), platelet factor-4 (B), fibrinogen (C), prothrombin fragment 1+2 (D), PAI-1 (E), D-dimer (F), non-esterified fatty acids (G), platelet activating factor acetylhydrolase activity in HDLs (H), and α1-antitrypsin in HDLs (I),. Results are shown as adjusted coefficients (percentage changes relative to baseline values) with 95% confidence intervals. HDL: high-density lipoprotein; MedDiet: traditional Mediterranean diet; MedDiet-Nuts: traditional Mediterranean diet enriched with mixed nuts; MedDiet-VOO: traditional Mediterranean diet enriched with virgin olive oil; PAF-AH: platelet activating factor acetylhydrolase.
Secondary observational analyses: associations of 1-year changes in MedDiet adherence, PA, and their combination with 1-year changes in atherothrombosis biomarkers
Each 1-point increment in MedDiet adherence was associated with less fibrinogen levels (−1.8%, 95% CI: −3.38; −0.17 –Figure 1C–). Each 100 METs·min/d increase in PA was linked to less prothrombin fragment 1+2 (−5.2%, 95% CI: −10.4; −0.077 –Figure 1D–) and less PAI-1 concentrations (−3.0%, 95% CI: −5.93; −0.12 –Figure 1E–). Finally, each 1-point increase in MedDiet adherence combined with an increment in 100 METs·min/d in PA was related to less D-dimer levels (−1.5%, 95% CI: −3.10; 0.056, P-value for interaction=0.047 – Figure 1F–). Exact coefficients are available in Supplemental Table 7.
DISCUSSION
Our results show that following a MedDiet and increasing PA practice are associated with benefits in several atherothrombosis biomarkers.
Atherothrombosis is strongly affected by inflammation and oxidative stress25,26. MedDiet is a lifestyle intervention known to improve these last two risk factors14, possibly explaining its protective effects on thrombosis as well. In our data, following the MedDiet-VOO intervention for 1 year improved HDL antithrombotic properties, whilst the MedDiet-Nuts decreased NEFA levels. We previously reported that the MedDiet-VOO intervention enhanced HDL functionality in the PREDIMED Study22. However, the present data extend this protection to anti-thrombotic HDL functions. The MedDiet antioxidant protection on HDL components could contribute to explaining the reported improved activity of HDL protective enzymes (due to a greater preservation of their non-oxidized structure)27. In addition, the MedDiet capacity to decrease plasma levels of inflammatory biomarkers may help explain the reduction in the concentrations of pro-inflammatory molecules bound to HDL particles16. In relation to NEFAs, they promote endothelial stress and platelet aggregation and are considered an emergent pro-thrombotic risk factor in high cardiovascular risk states such as obesity and hypertriglyceridemia9. As hypothetical mechanisms, polyunsaturated fatty acids (whose intake was particularly increased in the MedDiet-Nuts group) bind to GPR120 receptors in adipose tissue and decrease lipolysis, decreasing the circulating levels of free fatty acids28. In addition, MedDiet is a fiber-rich dietary pattern, and the intestinal metabolism of fiber leads to blood peaks of short-chain fatty acids (known to be able to bind other GPR family receptors with a similar anti-lipolytic effect29). Finally, increasing MedDiet adherence was also associated with lower fibrinogen levels, a biomarker known to be a connection point for inflammation and thrombosis5. This effect could also be partially explained by the MedDiet anti-inflammatory capacity14.
PA is another lifestyle factor related to improvements in oxidative stress and inflammation. This is due to its capacity to promote the expression of antioxidant and anti-inflammatory enzymes through the activation of AMP-activated protein kinase30. In our study, increases in PA were associated with decreases in prothrombin fragment 1+2 (directly proportional to the activation of prothrombin into thrombin) and PAI-1 concentrations. In addition, we also observed a synergistic protective effect between PA and MedDiet adherence on D-dimer levels. Pro-inflammatory states impair these three biomarkers31,32. Therefore, a decreased pro-inflammatory environment due to PA (alone or in combination with MedDiet) may help explaining these benefits.
Our study has limitations. First, study volunteers were elderly people at high cardiovascular risk, which does not allow the extrapolation of our findings to other populations. Second, we have only been able to report moderate changes or associations. However, this effect magnitude was expected because our work was based on modest real-life dietary modifications, the control diet was already a well-known healthy dietary pattern, and we did not provide any recommendation to increase PA. Finally, data on MedDiet adherence and PA levels were self-reported and could be somewhat imprecise. The main strengths of our study were its large sample size (N=358), randomized design, long-term duration (1 year), and strict quality control for laboratory biomarkers.
In conclusion, 1 year of intervention with MedDiet enhanced HDL antithrombotic properties and decreased NEFA levels in high cardiovascular risk individuals. The low-fat control intervention was the only dietary group linked to increases in the levels of platelet factor-4 and prothrombin activation relative to baseline. As observed in the secondary, observational analyses, increments in MedDiet adherence were associated with decreases in fibrinogen levels, increases in PA were linked to reductions in prothrombin fragment 1+2 and PAI-1 levels, and combined increases in MedDiet adherence and PA were related to improvements in D-dimer concentrations. As far as we know, this is the largest, most comprehensive, hypothesis-based analysis to date of the effects of healthy lifestyle on atherothrombosis mechanisms in high cardiovascular risk subjects. Our results support the improvement of thrombosis status after following a MedDiet or increasing PA. Further studies are needed to confirm whether these improvements mediate the reported cardioprotective benefits of such lifestyle modifications.
Data Availability
Considering that we do not have the explicit written consent of the study volunteers to yield their deidentified data at the conclusion of the study, individual participant data cannot be shared. The study protocol is available in the main publication of the study1 and in the study website (http://www.predimed.es/uploads/8/0/5/1/8051451/_1estr_protocol_olf.pdf), and a summary of the dietary intervention is also available in Supplemental Methods.
Sources of Funding
This work was supported by grants of Instituto de Salud Carlos III [OBN17PI02, PIE14/00045_INFLAMES, CB06/03/0019, CB06/03/0028, CD17/00122 (A.H.), and JR17/00022 (O.C.)], and Agència de Gestió d’Ajuts Universitaris i de Recerca (2017 SGR 222). CIBER de Fisiopatología de la Obesidad y Nutrición (CIBEROBN) is an initiative of the Instituto de Salud Carlos III, Madrid, Spain.
Disclosures
X.P. reports personal fees from Abbott, Esteve, Lacer, Rubio, and Sanofi. J.S.-S. reports to be a board member and grants from the International Nut and Dried Fruit Foundation; to be a member of the Executive Committee and personal fees from Instituto Danone Spain; personal fees from Aguas Font Vella Lanjarón, and Danone S.A; and grants from Eroski Distributors. F.A. reports personal fees from Menarini and AstraZeneca. E.R. reports personal fees and non-financial support from Merck, Sharp & Dohme; grants, personal fees, and non-financial support from the California Walnut Commission; grants and personal fees from Sanofi; personal fees and non-financial support from Danone; personal fees and non-financial support from Ferrer International; personal fees from Amarin; and grants from Pfizer. R.E. reports grants from Bicentury SA, Cerveza y Salud, Grand Fountaine, and Novartis SA; and personal fees from Brewers of Europe, FIVIN, Fundación Cerveza y Salud, Lilly Laboratories, and Wine and Culinary International Forum. The rest of the authors have nothing to disclose.
ACKNOWLEDGEMENTS
Authors wish to thank Daniel Muñoz-Aguayo, Gemma Blanchart, and Sònia Gaixas for their technical assistance.
Abbreviations
- HDL
- high-density lipoprotein
- MedDiet
- Traditional Mediterranean Diet
- MedDiet-Nuts
- Traditional Mediterranean Diet enriched in nuts
- MedDiet-VOO
- Traditional Mediterranean Diet enriched in virgin olive oil
- NEFA
- non-esterified fatty acids
- PA
- physical activity
- PAF-AH
- platelet activating factor-acetylhydrolase
- PAI1
- plasminogen activator inhibitor-1