Clinical Nutrition 40 (2021) 5674e5677 Contents lists available at ScienceDirect Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu Short Communication One year of omega 3 polyunsaturated fatty acid supplementation does not reduce circulating prothrombotic microvesicles in elderly subjects after suffering a myocardial infarction Gemma Chiva-Blanch a, b, e, g, *, Vibeke Bratseth a, g, Kristian Laake a, g, Harald Arnesen a, g, Svein Solheim a, g, Erik Berg Schmidt c, Lina Badimon d, f, Ingebjørg Seljeflot a, g a Center for Clinical Heart Research, Department of Cardiology, Oslo University Hospital Ullevål, 0450, Oslo, Norway b Department of Endocrinology and Nutrition, August Pi i Sunyer Biomedical Research Institute - IDIBAPS, Hospital Clínic of Barcelona, Barcelona, Spain c Department of Cardiology, Aalborg University Hospital, Denmark d Cardiovascular Program ICCC, Institut de Recerca Hospital Santa Creu i Sant Pau-IIB Sant Pau, Barcelona, Spain e n Biom Centro de Investigacio n (CIBEROBN), Instituto de Salud Carlos III (ISCIII), Madrid, Spain edica en Red Fisiopatología de la Obesidad y la Nutricio f n Biom Centro de Investigacio edica en Red Cardiovascular (CIBERCV), Instituto de Salud Carlos III (ISCIII), Madrid, Spain g Faculty of Medicine, University of Oslo, 0372, Oslo, Norway a r t i c l e i n f o s u m m a r y Article history: Background & aims: Circulating microvesicles (cMV) are both effectors and biomarkers of cardiovascular Received 23 August 2021 disease (CVD), and the effects of omega 3 polyunsaturated fatty acids (n3 PUFA) in MV shedding are not Accepted 6 October 2021 yet well known. Therefore, we aimed to investigate the effects of long-term n3 PUFA supplementation on cMV release from cells of the vascular compartment in elderly subjects at very high risk of CVD. Keywords: Methods: We included 156 elderly patients 2e8 weeks after suffering an acute myocardial infarction Microvesicles from the OMEMI cohort. Subjects were randomly allocated to receive 930 mg EPA þ 660 mg DHA (n3 Omega 3 fatty acids PUFA intervention) or corn oil (56% linoleic acid, 32% oleic acid, 10% palmitic acid) used as placebo daily EPA DHA for two years. At inclusion and after one-year follow-up, prothrombotic [annexin V (AV)þ] cMV derived Elderly from blood and vascular cells were phenotyped by flow cytometry. Thrombosis Results: No differences were observed in the levels of cMV between the randomized groups at inclusion in the study. After one-year follow-up, total AVþ, platelet-derived CD61þ/AVþ, and endothelial-derived CD31þ/AVþ and CD31þ/CD42b/AVþ cMV increased significantly in both groups. In the n3 PUFA sup- plemented group, platelet-derived CD62Pþ/AVþ, CD42bþ/AVþ and CD31þ/CD42bþ/AVþ; leukocyte- derived CD62Lþ/AVþ, CD45þ/AVþ, and CD11bþ/AVþ, as well as endothelial derived CD146þ/AVþ, CD62Eþ/AVþ, and CD309þ/AVþ cMV also increased significantly. No significant differences were however, observed in the changes of cMV levels between groups. Conclusion: In elderly Norwegians who have suffered a recent acute myocardial infarction and treated as per guidelines, long-term supplementation with 1.8 g/day n3 PUFA does not modulate prothrombotic MV release from blood and vascular cells. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01841944. © 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 1. Introduction to the bloodstream by almost all cell types. Increased concentra- tions of cMV have been observed in plasma from subjects with Circulating microvesicles (cMV) are a subset of large extracel- cardiovascular risk factors, atherosclerosis, and both in primary lular vesicles originated by cell membrane budding and released and secondary fatal and non-fatal cardiovascular events [1]. Abbreviations: cMV, circulating microvesicles; n3, omega 3; PUFA, polyunsaturated fatty acids. * Corresponding author. Endocrinology and Nutrition Department, Institut d'Investigacions Biome diques August Pi Sunyer (IDIBAPS), Hospital Clínic, Barcelona, Spain. E-mail address: gchiva@clinic.cat (G. Chiva-Blanch). https://doi.org/10.1016/j.clnu.2021.10.007 0261-5614/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). G. Chiva-Blanch, V. Bratseth, K. Laake et al. Clinical Nutrition 40 (2021) 5674e5677 Therefore, cMV are being considered a fingerprint of the onset and parental cells (blood and vascular cells) releasing the pro- progression of cardiovascular disease (CVD). Some cMV expose thrombotic cMV into the bloodstream, cMV were also characterized phosphatidylserine during their biogenesis, thus conferring about with markers of cell lineage and activation. Cell biomarkers used for 50- to 100-fold higher prothrombotic activity than platelets [2]. cMV phenotyping are listed in Supplementary Table 1. At baseline Whereas the mild effects of omega 3 polyunsaturated fatty acids and at the end of the study follow-up (2 years), fatty acid profile of (n3 PUFA) in platelet reactivity have been largely investigated, serum phospholipids were also measured as described in the data their involvement in secondary prevention of CVD, and more supplement of the reference [8]. specifically though prothrombotic MV shedding, has been a Statistical analyses were performed using the SPSS Statistical matter of debate. In fact, the studies performed reported contro- Analysis System version 23.0 (IBM, NY). Descriptive statistics versial or inconclusive results [3e7]. In many cases interventions [mean ± SD or n (%)] were used to describe the baseline charac- were short-termed (weeks), or the doses of the n3 PUFA tested teristics of the patients and the outcome variables. Changes in were too elevated to be sustained at the long term (around 5 g/ cMV after one-year follow-up were expressed as mean (95% con- day) [3], and only considered one or two specific MV phenotypes fidence interval). Normality of variables was assessed with the (of one or two cell origins). Moreover, all reported studies have ShapiroeWilk test, and all variables with a skewed distribution been performed in middle-aged subjects. Considering that cMV were transformed to their natural logarithms for parametric ana- exposing phosphatidylserine are prothrombotic, and the potential lyses and are shown as antilogarithms to facilitate interpretation of antithrombotic effects of n3 PUFA, we aimed to evaluate the ef- the results. Differences in baseline concentrations of cMV between fects of long-term n3 PUFA supplementation on prothrombotic groups of intervention were analyzed with the Student's t test for cMV release from cells of the vascular compartment in elderly unrelated samples. Changes in cMV after one-year follow-up within subjects with a recent acute myocardial infarction (AMI), and groups of intervention were analyzed with the Student's t test for thereby at very high risk of CVD. paired samples. Differences in changes in plasma concentrations of cMV after one-year follow-up between groups of intervention were 2. Materials and methods assessed by repeated measures ANOVA with the intervention group as the between subject factor. A two-tailed P-value of <0.05 was We included the first 156 patients of the “OMega-3 fatty acids considered statistically significant. in Elderly patients with Myocardial Infarction” (OMEMI; NCT01841944, Clinical-Trials.gov) cohort [8,9]. The protocol was 3. Results approved by the Regional Committee for Medical Research Ethics (ref. number: 2012/1422). The trial was conducted in Detailed baseline characteristics of the 156 subjects included in compliance with the declaration of Helsinki and with the rules this study can be found in Supplementary Table 2. Mean age was outlined in the guidelines for Good Clinical Practice. All partic- 74e75 years, approximately 75% of subjects were men and more ipants provided written informed consent. Patients aged 70e82 than 95% were under dual antiplatelet therapy. No differences in years old were recruited at Oslo University Hospital Ullevål, baseline characteristics were observed between groups of inter- Asker and Bærum Hospital, Vestre Viken and Akershus Univer- vention. After 2 years of intervention, the percentage of circulating sity Hospital Oslo, Norway, after a diagnosis of an AMI, and were EPA and DHA increased significantly at the expense of n6 PUFA in included in the study 2e8 weeks after hospitalization. Patients the intervention (n3 PUFA) group (Supplementary Table 3), indi- were treated per guidelines with standard medication. De- cating good compliance with the intervention. In addition, no dif- mographic data, clinical characteristics, cardiovascular history, ferences were observed in the levels of cMV between the groups at classic cardiovascular risk factors, and current pharmacological inclusion in the study (Supplementary Table 4). treatment were obtained from all subjects using a standardized As depicted in Table 1, after one-year of intervention, total AVþ, report form. At inclusion, patients were randomized in a 1:1 platelet-derived CD61þ/AVþ, and endothelial-derived CD31þ/AVþ ratio to receive either 1.8 g n-3 PUFA (930 mg EPA þ 660 mg and CD31þ/CD42b/AVþ cMV, increased in both groups. In the n3 DHA; Pikasol; Orkla Health, Oslo, Norway) or matching placebo PUFA group, platelet-derived CD62Pþ/AVþ, CD42bþ/AVþ, and (corn oil: 56% linoleic acid, 32% oleic acid, 10% palmitic acid) CD31þ/CD42bþ/AVþ; leukocyte-derived CD62Lþ/AVþ, CD45þ/AVþ, daily for 2 years. Subjects under n3 PUFA supplementation at and CD11bþ/AVþ; as well as endothelial derived CD146þ/AVþ, inclusion were instructed to limit n3 supplements to one child CD62Eþ/AVþ, and CD309þ/AVþ cMV also increased significantly. spoon of cod liver oil daily [8]. However, no significant differences were observed in changes in Blood samples were obtained after overnight fasting and pe- any cMV phenotype between n3 PUFA and the placebo after one- ripheral blood was withdrawn into 3.8% sodium citrate tubes for year of intervention. MV analysis, and into tubes for serum fatty acid quantification. There were also no significant associations between changes in Blood cells were removed by centrifuging whole blood 20 min at serum fatty acid concentrations and any cMV phenotype, also when 2500g at 4  C. Plasma was carefully recollected, leaving about exploring those with the most marked increase in EPA and/or DHA 0.2 cm of an undisturbed layer on top of the cells, and plasma and (data not shown). serum aliquots were immediately frozen and stored at 80  C until processing for isolation and quantification of cMV and fatty acids, 4. Discussion respectively. Additionally, biochemical and hematological param- eters were quantified with standardized routine methods. In our population of elderly Norwegians who have suffered an cMV analyses were performed between September and AMI within the last two months, daily supplementation with December 2015, and all researchers were blinded to the in- 1.8 g n3 PUFA for one year did not reduce MV shedding from cells of terventions. The randomization code was opened at the end of the blood and vascular compartment. In fact, after one year of n3 2020. Phosphatidylserine exposure in the surface of the MV was PUFA intervention, cMV from several phenotypes increased characterized by annexin V (AV) binding. Therefore, prothrombotic compared to baseline. Nevertheless, differences in these cMV (AVþ) cMV released by blood and vascular cells in plasma were concentrations did not achieve statistical significance compared to quantified at baseline and at one-year follow-up by flow cytometry the non-significant increases of cMV in the placebo group. Overall, as previously described [10]. To phenotypically characterize the these results show no effect of n3 PUFA supplementation on MV 5675 G. Chiva-Blanch, V. Bratseth, K. Laake et al. Clinical Nutrition 40 (2021) 5674e5677 Table 1 This study is not exempt of limitations. Corn oil, used as placebo, Differences in the concentration of cMV at baseline and at one year of intervention in is rich in n6 PUFA and may have potentially influenced the results. each study arm. The population studied was mainly composed of males (75%), and AVþ cMV n3 intervention (n ¼ 80) Placebo (n ¼ 76) P consequently, sex-specific differences observed in other studies [7] mean (95% CI) mean (95% CI) could not be analyzed. In addition, our study population was Total 286.52 (136.51, 436.52)a 222.22 (10.30, 434.14)a 0.471 Caucasian elderly subjects with a recent AMI treated as per CD142þ 3.80 (0.92, 8.52) 2.62 (3.90, 9.13) 0.706 guidelines, and were thus under double antiplatelet therapy. This CD61þ 236.15 (73.88, 398.42)a 235.79 (20.23, 451.34)a 0.781 guidelines-adhered best clinical practice may have rendered CD142þ/CD61þ 0.28 (1.63, 1.07) 0.80 (0.61, 2.22) 0.900 CD62Pþ 8.44 (1.88, 15.00)a 11.97 (1.21, 25.15) 0.688 platelets inactivated and may have abolished any effect of n3 sup- CD62Lþ 9.60 (1.82, 17.37)a 29.72 (16.17, 75.62) 0.626 plementation on prothrombotic MV shedding. The patients were CD146þ 1.05 (0.43, 1.67)a 0.67 (0.32, 1.65) 0.232 also allowed to continue with one child spoon of cod liver oil if they CD62Eþ 23.54 (2.29, 44.80)a 15.27 (1.25, 31.79) 0.508 used it before inclusion into the study, as previously discussed [8]. CD146þ/CD62Eþ 0.39 (0.02, 0.81) 0.19 (0.15, 0.52) 0.763 CD309þ 17.61 (2.86, 32.35)a 16.34 (8.44, 41.12) 0.192 Our findings may therefore have limited relevance to other pop- CD31þ 162.00 (60.32, 263.67)a 138.70 (14.57, 262.84)a 0.996 ulations. Also, the limit of quantification of the cytometer used is CD42bþ 103.89 (26.61, 181.18)a 75.37 (16.82, 167.57) 0.815 around 0.2 mm, and therefore smaller MV that could be modulated CD31þ/CD42bþ 88.92 (23.15, 154.68)a 67.24 (8.15, 142.63) 0.977 by the interventions were not quantified. CD31þ/CD42b- 72.34 (23.65, 121.02)a 71.53 (3.99, 139.08)a 0.776 CD235abþ 12.54 (14.37, 39.44) 9.34 (32.98, 14.30) 0.114 CD3þ/CD45þ 1.81 (3.95, 7.57) 0.89 (3.42, 5.19) 0.386 5. Conclusion CD45þ 15.24 (1.85, 28.63)a 7.82 (3.54, 19.17) 0.995 CD11bþ 2.80 (0.11, 5.49)a 0.76 (2.17, 3.69) 0.365 In this elderly Norwegian population who recently had suffered CD14þ 0.59 (3.02, 1.83) 0.90 (1.04, 2.84) 0.804 an AMI, one-year supplementation with n3 PUFA did not have any CD11bþ/CD14þ 0.84 (1.87, 0.19) 0.26 (0.47, 1.00) 0.457 CD142þ/CD14þ 1.87 (3.85, 0.11) 0.18 (0.88, 1.24) 0.227 effect on prothrombotic MV release to the bloodstream. Thus, our results do not support that supplementation with omega 3 PUFA Results are expressed as mean (95% confidence interval) differences between before and after each intervention for each cMV phenotype (quantified as cMV/mL plasma). reduce thrombotic risk through a MV-mediated mechanism in P from the comparison of the differences between interventions after 1 year elderly subjects under dual antiplatelet therapy. (repeated measures ANOVA with the intervention group as the between subject factor). a Funding statement Significant changes after one year of intervention (Student's t test for paired samples). This work was supported by the Stein Erik Hagens Foundation of Clinical Heart Research, Oslo, Norway. release. This was despite satisfactory compliance to the study drug, as indicated by significant increases in serum levels of the relevant Author contributions fatty acids. These results fit with the main results from the OMEMI trial [8], in which n3 PUFA supplementation (EPA þ DHA) did not Conceptualization, G.C.-B., L.B., and I.S.; data curation, K.L., V.B, have a significant impact on secondary cardiovascular events. and S.S.; formal analysis, G.C.-B., E.B-S, and V.B.; funding acquisi- Importantly, these results endorse the potential role of cMV as tion, H.A, and I.S.; investigation, G.C.-B., and V.B.; methodology, biomarkers of progression and severity of CVD, and as biomarkers G.C.-B. and V.B.; resources, H.A., S.S., and I.S.; software, G.C.-B.; of the effect of diet on CVD. supervision, H.A.; writingdoriginal draft, G.C.-B.; writingdreview The effects of n3 PUFA supplementation on MV release have and editing, L.B., and I.S; final approval: all authors. been scarcely investigated. In a study performed in 46 subjects below 75 years, prothrombotic platelet-derived CD61þ/AVþ and Conflict of Interest monocyte-derived CD14þ/AVþ cMV concentrations decreased after the administration of 5.2 g/day of n3 PUFA for 12 weeks, compared The authors declare no conflicts of interest. to olive oil [3]. However, this study was performed in younger subjects after an AMI, with a short-term intervention with almost Acknowledgements 3-fold the dose administered in our study. In 94 healthy middle- aged subjects, neither supplementation with EPA, nor with DHA We are indebted to the staff and coworkers at Center for Clinical for 4 weeks (1.2 g/day each intervention), did modulate CD36þ cMV Heart Research Ullevål, Oslo University Hospital; Asker and Bærum release [6]. However, in 84 moderate cardiovascular risk subjects Hospital, Vestre Viken; Akershus University Hospital and The Lipid aged 21e65 years, 1.5 g/day of n3 PUFA decreased the circulating Research Laboratory, Aalborg University Hospital for their invalu- levels of endothelial-derived CD31þ/CD42b but not platelet- able contribution. derived CD31þ/CD42bþ MV [4]. Supplementation with 1.8 g/day of EPA only during 6 months, decreased the release of pro- thrombotic endothelial-derived CD51þ/AVþ cMV in 126 diabetic Appendix A. Supplementary data and hyperlipidemic individuals, but not in non-diabetic hyper- lipidemic subjects [5]. The disparities observed in the different Supplementary data to this article can be found online at studies, overall indicate that the dose of n3 PUFA supplementation https://doi.org/10.1016/j.clnu.2021.10.007. might be crucial for the antithrombotic activity of the n3 PUFA, as for instance, 4 g of icosapent ethyl reduced the risk of ischemic References events in the Reduce-IT study [11], whereas 1.8 g n3 PUFA did not in [1] Badimon L, Suades R, Vilella-Figuerola A, Crespo J, Vilahur G, Escate R, et al. the OMEMI cohort [8]. Additionally, the proportion of EPA and DHA Liquid biopsies: microvesicles in cardiovascular disease. Antioxid Redox Signal in the supplements, as well as the age and healthy status of the 2020;33:645e62. https://doi.org/10.1089/ars.2019.7922. subjects, including use of medications such as antiplatelet agents, [2] Sinauridze EI, Kireev DA, Popenko NY, Pichugin AV, Panteleev MA, Krymskaya OV, et al. Platelet microparticle membranes have 50- to 100-fold may significantly determine the effects of n3 PUFA on cell activa- higher specific procoagulant activity than activated platelets. Thromb Hae- tion and cMV release. mostasis 2007;97:425e34. 5676 G. Chiva-Blanch, V. Bratseth, K. Laake et al. Clinical Nutrition 40 (2021) 5674e5677 [3] Turco SD, Basta G, Lazzerini G, Evangelista M, Rainaldi G, Tanganelli P, et al. J Nutr Biochem 2012;23:1128e33. https://doi.org/10.1016/j.jnutbio.2011.06. Effect of the administration of n-3 polyunsaturated fatty acids on circulating 006. levels of microparticles in patients with a previous myocardial infarction. [8] Kalstad AA, Myhre PL, Laake K, Tveit SH, Schmidt EB, Smith P, et al. Effects of Haematologica 2008;93:892e9. https://doi.org/10.3324/haematol.11845. n-3 fatty acid supplements in elderly patients after myocardial infarction: a [4] Wu S-Y, Mayneris-Perxachs J, Lovegrove JA, Todd S, Yaqoob P. Fish-oil sup- randomized, controlled trial. Circulation 2021;143:528e39. https://doi.org/ plementation alters numbers of circulating endothelial progenitor cells and 10.1161/CIRCULATIONAHA.120.052209. microparticles independently of eNOS genotype. Am J Clin Nutr 2014;100: [9] Laake K, Myhre P, Nordby LM, Seljeflot I, Abdelnoor M, Smith P, et al. Effects of 1232e43. https://doi.org/10.3945/ajcn.114.088880. omega 3 supplementation in elderly patients with acute myocardial infarc- [5] Nomura S, Shouzu A, Omoto S, Inami N, Ueba T, Urase F, et al. Effects of tion: design of a prospective randomized placebo controlled study. BMC eicosapentaenoic acid on endothelial cell-derived microparticles, angiopoie- Geriatr 2014;14:74. https://doi.org/10.1186/1471-2318-14-74. tins and adiponectin in patients with type 2 diabetes. J Atherosclerosis [10] Chiva-Blanch G, Laake K, Myhre P, Seljeflot I, Chiva-Blanch G, Laake K, et al. Thromb 2009;16:83e90. High adherence to the Nordic diet is associated with lower levels of total and [6] Phang M, Thorne RF, Alkhatatbeh MJ, Garg ML, Lincz LF. Circulating CD36þ platelet-derived circulating microvesicles in a Norwegian population. Nutri- microparticles are not altered by docosahexaenoic or eicosapentaenoic acid ents 2019;11:1114. https://doi.org/10.3390/NU11051114. supplementation. Nutr Metab Cardiovasc Dis 2016;26:254e60. https:// [11] Bhatt DL, Steg PG, Miller M, Brinton EA, Jacobson TA, Ketchum SB, et al. doi.org/10.1016/j.numecd.2015.12.003. Cardiovascular risk reduction with icosapent ethyl for hyper- [7] Phang M, Lincz L, Seldon M, Garg ML. Acute supplementation with eicosa- triglyceridemia. N Engl J Med 2019;380:11e22. https://doi.org/10.1056/ pentaenoic acid reduces platelet microparticle activity in healthy subjects. nejmoa1812792. 5677