Heart disease in women and men
van der Ende, Maaike Yldau
DOI:10.33612/diss.103508645
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
van der Ende, M. Y. (2019). Heart disease in women and men: insights from Big Data. Rijksuniversiteit Groningen. https://doi.org/10.33612/diss.103508645
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
INSIGHTS FROM BIG DATA
Cover design: Design Your Thesis | www.designyourthesis.com Layout: Design Your Thesis | www.designyourthesis.com Print: Ridderprint | www.ridderprint.nl
ISBN: 978-94-6375-609-9
Financial support by the following sponsor for the publication of this thesis is gratefully acknowledged: Rijksuniversiteit Groningen, Groningen University Institute for Drug Exploration (GUIDE).
Copyright © 2019 by Maaike Yldau van der Ende. All rights reserved. Any unauthorized reprint or use of this material is prohibited. No part of this thesis may be reproduced, stored or transmitted in any form or by any means, without written permission of the author or, when appropriate, of the publishers of the publications.
Insights from Big Data
Proefschrift
ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen
op gezag van de
rector magnificus prof. dr. C. Wijmenga en volgens besluit van het College voor Promoties.
De openbare verdediging zal plaatsvinden op woensdag 18 december 2019 om 14:30 uur
door
Maaike Yldau van der Ende
geboren op 10 maart 1993Prof. dr. H. Snieder Copromotor Dr. E. Lipsic Beoordelingscommissie Prof. dr. B.H.Ch. Stricker Prof. dr. J.L. Hillege Prof. dr. F.W. Asselbergs
Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowledged.
Chapter 1. General introduction 9
I. PREVALENCE AND PREVENTION OF CARDIOVASCULAR DISEASE AND RISK FACTORS
Chapter 2. The Lifelines cohort study: prevalence and treatment ofcardiovascular disease and risk factors
23
Chapter 3. The effect of feedback on cardiovascular risk factors on optimization of primary prevention: a population based cohort study
43
Chapter 4. Population-based values and abnormalities of the
electrocardiogram in the general Dutch population: the Lifelines cohort study
61
Chapter 5. Resting heart rate and coronary artery disease: a genome-wide meta-analysis and Mendelian randomization analysis
85
II. UNRECOGNIZED MYOCARDIAL INFARCTION
Chapter 6. Prevalence of electrocardiographic unrecognized myocardial infarction and its association with mortality
107
Chapter 8. The Groningen electrocardiographic criteria for left ventricular hypertrophy: a sex-specific analysis
149
Chapter 9. Causal pathways from blood pressure to larger QRS amplitudes: a Mendelian randomization study
169
Chapter 10. Effect of systolic blood pressure on left ventricular structure and function: a Mendelian randomization study
189
Chapter 11. General discussion and future perspectives 211
Chapter 12. Nederlandse samenvatting 225
Appendix. Dankwoord 237
Curriculum Vitae 241
EPIDEMIOLOGY OF CARDIOVASCULAR DISEASE AND RISK FACTORS
Cardiovascular disease and mortality
In 2016, the worldwide number of people who died from cardiovascular disease (CVD) was around 17.9 million, representing 31% of all global deaths1. From 1990 to 2013,
population aging (55% increase from 1990) and population growth (25% increase from 1990) resulted in a 40% increase in the number of CVD deaths, despite an overall decrease in age-specific death rates2. Ischemic heart disease was the largest contributor
to this increase in the number of CVD deaths (2.4 million of the overall increase of 5.0 million CVD deaths). Central and Western Europe were the only regions with a decrease in CVD deaths between 1990 and 2013, with declines of 5.2% and 12.8%, respectively2.
Small relative gains in health, paired with a lower degree of population aging, resulted in this decline in total CVD deaths in Western Europe. In Europe, the number of deaths due to CVD was higher for women (2.1 million) than in men (1.7 million)3. CVD also
accounted for a larger proportion of all deaths in women (49%) than in men (40%)3.
Focusing on the Netherlands, sex differences in CVD mortality can be seen as well. In 2017, 20.039 women died because of CVD compared to 18.080 men, which account for 25% and 26% of the total mortality numbers4. Due to the higher number of women
with an older age in the Netherlands, more women than men are dying from CVD in the Netherlands nowadays. From 1980, there is a decline in CVD mortality rate of 76% in men and 74% in women in the Netherlands5. However, this fall in CVD mortality is
not going to persist6 and is already slowing or even plateaued for young adults (<55
years)5. Also, sex-differences in clinical and procedural characteristics and outcome
in individuals with ischemic heart disease are described in the Netherlands7. Women
with ischemic heart disease more often have hypertension and diabetes compared to men, but have less extensive coronary artery disease and undergo less often coronary angiography. One-year mortality after ischemic heart disease is higher in women than in men7.
Cardiovascular risk factors and health behavior
Controlling cardiovascular risk factors together with the improvements in treatment of CVD resulted in a decline in CVD deaths since the early 1970s in the Netherlands5.
However, there is still an expected rise in cardiovascular risk factors such as obesity and diabetes by 20308,9. Important differences in the occurrence of cardiovascular risk factors
are identified between European high-income and middle-income countries. For both men and women, the prevalence of hypertension is lower in high-income countries compared to middle-income countries. Smoking prevalence in men (but not in women)
is also lower in high-income countries compared to middle-income countries3. In the
Netherlands, a large percentage of the adult population between 18 and 75 years is at risk for developing CVD since around 1.9 million, 2 million, 3 million and 2.5 million of these inhabitants have obesity, elevated cholesterol levels, hypertension or smoke, respectively4. The effect of these risk factors on CVD outcomes are largely similar
between women and men, however prolonged smoking is significantly more hazardous for women than for men10.
Besides these classical cardiovascular risk factors, several new risk factors have been identified in an effort to improve risk assessment for CVD. Resting heart rate is one of these potential cardiovascular risk factors of particular interest11. Elevated resting
heart rate has been associated with higher risk of CVD and death in observational studies12. However, this association does not provide sufficient evidence for a shared or
causal relationship. Additionally, during the last years, major developments in CVD risk prediction during the last years have been made by the identification of many single-nucleotide polymorphisms (SNPs) that are associated with cardiovascular risk factors or events. Based on these SNPs, genetic risk scores can be generated for improving (causal) risk prediction of CVD.
To conclude, the worldwide and nationwide burden of CVD is high, especially in women. The decline in CVD death is plateauing and the number of individuals with cardiovascular risk factors is high and expecting to rise. Primary prevention, in terms of lifestyle intervention and preventive medication, remains therefore an important target to reduce the incidence of CVD. Novel technologies, like genetic analyses, are of major importance for identifying individuals at risk and for determining causal relationships between (novel) risk factors and CVD.
THE ELECTROCARDIOGRAM
The electrocardiogram (ECG) is an important tool for diagnosing CVD. The first step in the development of the ECG dated back to 1786 with the recording of an electrical current from skeletal muscles. In 1842, an electrical current was described that accompanied every heart beat in a frog. The first human ECG was reported by Augustus Waller in 1887. He showed that the electrical activity preceded ventricular contraction13. Willem
Einthoven was able to demonstrate the ECG in five deflections, nowadays known as the P wave, QRS complex and T wave and was awarded the Nobel Prize in physiology for the invention of the ECG in 192413. Since that time, many new technologies have
ECG is the most important test for interpretation of the cardiac rhythm, conduction system abnormalities and the detection of myocardial ischemia. The ECG is also of great value in the evaluation of other types of cardiac abnormalities including valvular heart disease, cardiomyopathy, pericarditis, and left ventricular hypertrophy (LVH). Finally, the ECG can be used to monitor drug treatment (specifically antiarrhythmic therapy) and to detect metabolic disturbances such as alternations in serum calcium and potassium concentrations.
Since 1960, is has been shown that ECG parameters differ between sexes14. For example,
PR interval and QRS duration are longer in men compared to women15. In addition,
men have higher Q, R and S wave amplitudes than women15. Therefore, it is crucial that
detecting cardiac disorders with the ECG, such as LVH, takes these sex differences into account.
(Unrecognized) myocardial infarction and the electrocardiogram
The ECG has a major role in diagnosing MI. ST-elevations on the ECG are highly sensitive for acute ST-elevation MI (STEMI) and have a leading role in determining further treatment16. Individuals with complains of angina and ST-elevations on the ECG should
undergo percutaneous coronary intervention as soon as possible without the need for further diagnostic exploration16. In addition, the ECG is quick in diagnosing STEMI,
available in ambulances all over the world, which makes it possible to diagnose STEMI before a patient reaches the hospital. The enormous decline of in-hospital mortality rates after STEMI (30% in 1950 to 4% in 2010)is therefore, in addition to the development of the critical care unit and reperfusion therapy, partly due to the possibility of quickly diagnosing of STEMI by the ECG17.
Although major steps have been made in diagnosis and treatment of recognized MI, a large number of MIs remain unrecognized. These individuals with unrecognized MI are not treated at all. Earlier studies reported that 22% to 64% of the patients with coronary artery disease experience an unrecognized MI, with atypical or no symptoms of MI at all18. These patients do not receive secondary prevention and are at increased risk of
clinical CVD compared to individuals without previous MI19,20,21 and even compared
to individuals in whom MI was recognized22. In addition, unrecognized MI has been
associated with an increased risk for all-cause mortality23. It has been described that
MIs in women are more associated with atypical symptoms compared to men, and are therefore at higher risk for remaining unrecognized24. Examples of atypical symptoms
are nausea, dizziness and fatigue. With the help of the ECG there is a possibility to screen for individuals with unrecognized MI. The most prevalent sign of an old MI on the ECG is a pathological Q wave. This pathological Q wave can be seen as an electrical
hole in the heart, since scar tissue, developed as a result of a MI, is electrically dead. Diagnosing Q wave MI requires that pathological Q waves are present in at least two anatomically contiguous leads25. Large population based cohort studies are set up to
evaluate the epidemiology of CVD and cardiovascular risk factors. Cohort studies, with available 12-lead ECGs, give the opportunity to search for unrecognized MI and related sex-differences in the general population. It is known that the sensitivity of the ECG for diagnosing old MIs is low and lies between 0.30 and 0.5826. However, due to the
non-invasiveness and cost-effectiveness, the ECG remains an important tool for detecting unrecognized MI.
Left ventricular hypertrophy and the electrocardiogram
LVH is a marker of the pathophysiologic response of the myocardium to chronic pressure or volume overload and is associated with future cardiovascular events27.
Major risk factors of LVH are hypertension and aortic stenosis. Commonly used imaging techniques for detecting LVH are echocardiography and cardiac magnetic resonance imaging (MRI). Studies using echocardiographic criteria for detecting LVH reported a prevalence of 15% of LVH in men and 9% in women in the general population28.
Besides imaging techniques, the ECG is a widely-used tool for detecting LVH. Increased left ventricular mass can be identified by the ECG according to high Q, R or S wave amplitudes and long QRS duration. So far, many ECG-LVH criteria are developed with different sums of amplitudes of different leads of the ECG, but these criteria show low sensitivity for detecting LVH as compared to echocardiography. The accuracy of the ECG for diagnosing LVH has been described to be less in women than in men29, but LVH
detected by the ECG is a stronger risk factor for incident CVD events in women than in men30. Antihypertensive treatment can decrease LVH and improve left ventricular
dysfunction31. The higher chance of false negative findings of LVH in women may
therefore lead to a higher rate of undertreatment of LVH and related cardiovascular events32. Existing ECG-LVH criteria are similar in men and women and most of these
criteria do not have sex-specific cut-off points, suggesting that there is still some room for improvement.
To conclude, the ECG plays a central role in diagnosing CVD in the acute phase, but is also an important tool for detecting (unrecognized) cardiac disorders on population level. It is important to take sex-differences, which manifest on the ECG, into account.
THIS THESIS
Big data and study designs of this thesis
The increasing generation and development of digital data and computational science make it possible to gain new insights from enormous data sets, known as big data. In this thesis, big data of two large cohort studies are used: the Lifelines cohort study and the UK Biobank. In addition, data of the IADB.nl pharmacy database was used. The Lifelines cohort study is a contemporary population-based study, which was initiated in 2006 to improve our knowledge on (cardiovascular) healthy ageing33. The Lifelines
cohort study includes over 167,000 individuals of the three northern provinces of the Netherlands and aims to follow them for 30 years. The PharmLines Initiative was started in 2017 to link data of the Lifelines cohort study to the University Groningen prescription database, IADB.nl. The IADB.nl database is a pharmacy database that contains prescription data from approximately 600,000 patients in the northern part of the Netherlands34. At the beginning of 2019, around 60,000 of the participants of
the Lifelines cohort study, could be linked to the IADB.nl database. The UK Biobank is a population based prospective study established for investigating genetic and non-genetic determinants of diseases. Between 2006 and 2010, over 500,000 participants of the United Kingdom aged between 40 and 69 years were recruited35. Imaging visits of
the UK Biobank were initiated in 2015. During these visits, cardiac magnetic resonance imaging (CMR) was performed, with the aim to scan over 100.000 participants36.
Baseline characteristics of participants of these cohort studies were used in sectional study designs to determine the prevalence of CVD and risk factors. A cross-sectional study design was also used for testing the accuracy, in terms of sensitivity and specificity, of diagnostic criteria for CVD. Follow-up data was used in longitudinal analyses to obtain incidence rates of CVD and to determine the predictors of incident CVD. Nested matched case-control designs were used for studies in which was focused on one specific CVD. The generation of a matched control group out of the total study population, enabled a comparison of risk factors among cases and controls with less confounding factors. Also, the use of nested case-control designs resulted in smaller sample sizes and enabled additional measurements in these individuals.
In the remaining chapters of this thesis, summary statistics of performed genome wide association studies (GWAS) on CVD and cardiovascular risk factors were used in Mendelian randomization (MR) analyses. Also, a GWAS meta-analysis was performed and described in one of the chapters of this thesis. MR analyses are designed to investigate the causal nature of the relationship between risk factors and outcomes
in observational data in the presence of confounding factors37. Using genetic variants
as instruments, which are randomly assigned when passed from parents to offspring during meiosis, the genotype distribution in the population should be unrelated to the presence of confounders.
Aim of this thesis
The first aim of this thesis is to evaluate the prevalence and pharmaceutical prevention of CVD and risk factors in the general population, especially the northern part of the Netherlands. Secondly, it is aimed to evaluate the occurrence of (unrecognized) MI detected by the ECG in women and men of the general population and to determine the predictors, related symptoms and association with mortality of (unrecognized) MI. Lastly, the performance of ECG criteria for detecting LVH is studied and these criteria are used for determining causal relationships between cardiovascular risk factors and LVH and between LVH and mortality and longevity.
Part I: Prevalence and prevention of cardiovascular disease and risk factors
In part I, occurrence of CVD and risk in the general population are investigated. Due to changes in lifestyle and the ageing population, the prevalence of CVD and related healthcare costs are expected to increase. Epidemiologic studies are crucial to improve our understanding of the genetic, behavioral and environmental determinants associated with CVD and its risk factors. Epidemiologic studies in the past, including the Framingham Heart study initiated in 1948, have contributed enormously to our understanding of CVD and its risk factors. However, advances in treatment as well as changes in behavior and lifestyle have occurred and contemporary data is needed for determining the occurrence, pharmaceutical prevention and associates of CVD nowadays. A contemporary population-based cohort study is the Lifelines cohort study33. In chapter 2 the prevalence of CVD, its risk factors and utilization of primary
prevention by drug treatment in the Lifelines cohort study are presented. In chapter
3, the use of cardiovascular preventive medication is determined in individuals at risk
for CVD (aged <70 years) based on the “European Society of Cardiology” (ESC) and “The Dutch College of General Practitioners” (Nederlands Huisartsen Genootschap (NHG)) guidelines using the PharmLines data. After the baseline visit of Lifelines, all participants and their general practitioners are informed about their individual health status, including their cardiovascular risk. Cardiovascular preventive medication usage is reported before and after the Lifelines baseline visit. In chapter 4, an overview is given about sex-specific population-based values of heart rate, P wave and QRS complex duration, PQ and QTc interval and the P, QRS and T axis measured by ECG. Additionally, the prevalence of abnormalities on the ECG in the Lifelines population free of CVD is
reported. In chapter 5, resting heart rate as a new potential cardiovascular risk factor is evaluated. Genetic variants associated with resting heart rate are determined in a GWAS meta-analysis of 835,465 individuals including 100 cohorts, and are used to test for a causal relationship with coronary artery disease and MI in MR analyses.
Part II: Unrecognized myocardial infarction
In chapters 6 and 7, individuals who have ECG signs of MI but are unaware of this are studied. In chapter 6, the prevalence of unrecognized MI in 152,124 individuals of the Lifelines cohort study is determined. During the baseline visit, a 12-lead ECG was made of all participants. A comparison is made between individuals with unrecognized MI and individuals with recognized MI and a generated matched control group. Risk factors related to unrecognized MI are determined and analyses are performed to determine whether a relationship exists between unrecognized MI and mortality. In chapter 7, incidence rate, symptomatology and predictors of unrecognized MI are described. 57,276 women and 39,927 men of the Lifelines cohort study with available baseline and follow-up ECG are included in this study. An incident unrecognized MI is defined when a participant has ECG signs corresponding to MI during follow-up in absence of a reported history of MI and pathologic ECG signs at the baseline ECG. Sex-differences in incidence of unrecognized MI are described, as well as the sex-specific predictors and symptomatology of unrecognized MI.
Part III: Left ventricular hypertrophy
In chapter 8, the accuracy of existing ECG voltage criteria for predicting LVH is described in 1,670 men and 1,962 women of the UK Biobank study with available CMR and 12-lead ECG data. In addition, the first sex-specific ECG criteria for detecting LVH using CMR are developed. In chapter 9, MR analyses are performed using summary statistics of GWAS to determine which cardiovascular risk factors causally lead to lager QRS amplitudes or longer QRS duration. Also, it is aimed to determine whether genetically predicted larger QRS amplitudes or longer QRS duration have a causal relationship with mortality and longevity. Finally, in chapter 10, MR analyses are performed to investigate whether causal relationships exist between systolic blood pressure and left ventricular structure and function, as determined by CMR.
REFERENCES
1. World Health Organization. The top 10 causes of death. Available at: https://www.who.int/
news-room/fact-sheets/detail/the-top-10-causes-of-death. Updated 2018. Assessed 2019.
2. Roth GA, Forouzanfar MH, Moran AE, et al. Demographic and epidemiologic drivers of global
cardiovascular mortality. N Engl J Med. 2015; 372: 1333-1341.
3. Timmis A, Townsend N, Gale C et al. European Society of Cardiology: Cardiovascular Disease
Statistics 2017. Eur Heart J. 2018; 39: 508-579.
4. de Boer AR, van Dis I, Visseren FLJ, Vaartjes I, Bots ML. Hart- en vaatziekten in Nederland 2018, cijfers over risicofactoren, hartinterventies, ziekte en sterfte. Den Haag: Hartstichting, 2018.
5. Vaartjes I, O’Flaherty M, Grobbee DE, Bots ML, Capewell S. Coronary heart disease mortality
trends in the Netherlands 1972-2007. Heart. 2011; 97: 569-573.
6. O’Flaherty M, Buchan I, Capewell S. Contributions of treatment and lifestyle to declining CVD
mortality: why have CVD mortality rates declined so much since the 1960s? Heart. 2013; 99: 159-162.
7. Ten Haaf ME, Bax M, Ten Berg JM, et al. Sex differences in characteristics and outcome in acute coronary syndrome patients in the Netherlands. Neth Heart J. 2019; 27: 263-271.
8. Koopman C, Vaartjes I, Blokstra A, et al. Trends in risk factors for coronary heart disease in the
Netherlands. BMC Public Health. 2016; 16: 835-016.
9. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics-2018 Update:
A Report From the American Heart Association. Circulation. 2018; 137: e67-e492.
10. Appelman Y, van Rijn BB, Ten Haaf ME, Boersma E, Peters SA. Sex differences in cardiovascular risk factors and disease prevention. Atherosclerosis. 2015; 241: 211-218.
11. Thomas MR, Lip GY. Novel Risk Markers and Risk Assessments for Cardiovascular Disease. Circ
Res. 2017; 120: 133-149.
12. Seviiri M, Lynch BM, Hodge AM, et al. Resting heart rate, temporal changes in resting heart rate, and overall and cause-specific mortality. Heart. 2018; 104: 1076-1085.
13. AlGhatrif M, Lindsay J. A brief review: history to understand fundamentals of electrocardiography. J Community Hosp Intern Med Perspect. 2012; 2: 10.3402/jchimp. v2i1.14383. Print 2012.
14. Simonson E, Blackburn H, Puchner TC, Eisenberg P, Ribeiro F, Meja M. Sex Differences in the Electrocardiogram Circulation. 1960; 598-601.
15. Rijnbeek PR, van Herpen G, Bots ML, et al. Normal values of the electrocardiogram for ages 16-90 years. J Electrocardiol. 2014; 47: 914-921.
16. Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Rev Esp Cardiol (Engl
Ed). 2017; 70: 1082.
17. Braunwald E. The treatment of acute myocardial infarction: the Past, the Present, and the Future. Eur Heart J Acute Cardiovasc Care. 2012; 1: 9-12.
18. Valensi P, Lorgis L, Cottin Y. Prevalence, incidence, predictive factors and prognosis of silent myocardial infarction: a review of the literature. Arch Cardiovasc Dis. 2011; 104: 178-188. 19. Ikram MA, Hollander M, Bos MJ, et al. Unrecognized myocardial infarction and the risk of
stroke: the Rotterdam Study. Neurology. 2006; 67: 1635-1639.
20. Krijthe BP, Leening MJ, Heeringa J, et al. Unrecognized myocardial infarction and risk of atrial fibrillation: the Rotterdam Study. Int J Cardiol. 2013; 168: 1453-1457.
21. Qureshi WT, Zhang ZM, Chang PP, et al. Silent Myocardial Infarction and Long-Term Risk of Heart Failure: The ARIC Study. J Am Coll Cardiol. 2018; 71: 1-8.
22. Yano K, MacLean CJ. The incidence and prognosis of unrecognized myocardial infarction in the Honolulu, Hawaii, Heart Program. Arch Intern Med. 1989; 149: 1528-1532.
23. Zhang ZM, Rautaharju PM, Prineas RJ, et al. Race and Sex Differences in the Incidence and Prognostic Significance of Silent Myocardial Infarction in the Atherosclerosis Risk in Communities (ARIC) Study. Circulation. 2016; 133: 2141-2148.
24. Mehta LS, Beckie TM, DeVon HA, et al. Acute Myocardial Infarction in Women: A Scientific Statement From the American Heart Association. Circulation. 2016; 133: 916-947.
25. White H, Thygesen K, Alpert JS, Jaffe A. Universal MI definition update for cardiovascular disease. Curr Cardiol Rep. 2014; 16: 492-014.
26. Sandler LL, Pinnow EE, Lindsay J. The accuracy of electrocardiographic Q waves for the detection of prior myocardial infarction as assessed by a novel standard of reference. Clin
Cardiol. 2004; 27: 97-100.
27. Porthan K, Kentta T, Niiranen TJ, et al. ECG left ventricular hypertrophy as a risk predictor of sudden cardiac death. Int J Cardiol. 2019; 276: 125-129.
28. Schirmer H, Lunde P, Rasmussen K. Prevalence of left ventricular hypertrophy in a general population; The Tromso Study. Eur Heart J. 1999; 20: 429-438.
29. Okin PM, Roman MJ, Devereux RB, Borer JS, Kligfield P. Electrocardiographic diagnosis of left ventricular hypertrophy by the time-voltage integral of the QRS complex. J Am Coll Cardiol. 1994; 23: 133-140.
30. Porthan K, Niiranen TJ, Varis J, et al. ECG left ventricular hypertrophy is a stronger risk factor for incident cardiovascular events in women than in men in the general population. J
Hypertens. 2015; 33: 1284-1290.
31. Wachtell K, Bella JN, Rokkedal J, et al. Change in diastolic left ventricular filling after one year of antihypertensive treatment: The Losartan Intervention For Endpoint Reduction in Hypertension (LIFE) Study. Circulation. 2002; 105: 1071-1076.
32. Ghali JK, Liao Y, Simmons B, Castaner A, Cao G, Cooper RS. The prognostic role of left ventricular hypertrophy in patients with or without coronary artery disease. Ann Intern Med. 1992; 117: 831-836.
33. Scholtens S, Smidt N, Swertz MA, et al. Cohort Profile: LifeLines, a three-generation cohort study and biobank. Int J Epidemiol. 2014; .
34. Visser ST, Schuiling-Veninga CC, Bos JH, de Jong-van den Berg, L T, Postma MJ. The population-based prescription database IADB.nl: its development, usefulness in outcomes research and challenges. Expert Rev Pharmacoecon Outcomes Res. 2013; 13: 285-292.
35. Sudlow C, Gallacher J, Allen N, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015; 12: e1001779.
36. Petersen SE, Matthews PM, Francis JM, et al. UK Biobank’s cardiovascular magnetic resonance protocol. J Cardiovasc Magn Reson. 2016; 18: 8-016.
37. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015; 44: 512-525.
Prevalence and treatment of cardiovascular disease
in the northern part of the Netherlands
• • •
M. Yldau van der Ende*, Minke H.T. Hartman*, Yanick Hagemeijer, Laura Meems,
Hendrik Sierd de Vries, Ronald P. Stolk, Rudolf A. de Boer, Anna Sijtsma, Peter van der Meer, Michiel Rienstra, Pim van der Harst. *Authors M. Yldau van der
Ende and Minke H.T. Hartman contributed equally to the current study.
ABSTRACT
Background
The LifeLines cohort study is a large three-generation prospective study and biobank. Recruitment and data collection started in 2006 and follow-up is planned for 30 years. The central aim of LifeLines is to understand healthy ageing in the 21st century. Here, the
study design, methods, baseline and major cardiovascular phenotypes of the LifeLines cohort study are presented.
Methods and Results
Baseline cardiovascular phenotypes were defined in 9,700 juvenile (8–18 years) and 152,180 adult (≥18 years) participants. Cardiovascular disease (CVD) was defined using ICD-10 criteria. At least one cardiovascular risk factor was present in 73% of the adult participants. The prevalence, adjusted for the Dutch population, was determined for risk factors (hypertension (33%), hypercholesterolemia (19%), diabetes (4%), overweight (56%), and current smoking (19%)) and CVD (myocardial infarction (1.8%), heart failure (1.0%), and atrial fibrillation (1.3%)). Overall CVD prevalence increased with age from 9% in participants <65 years to 28% in participants ≥65 years. Of the participants with hypertension, hypercholesterolemia and diabetes, respectively 75%, 96% and 41% did not receive preventive pharmacotherapy.
Conclusions
The contemporary LifeLines cohort study provides researchers with unique and novel opportunities to study environmental, phenotypic, and genetic risk factors for CVD and is expected to improve our knowledge on healthy ageing. In this contemporary cohort, we identified a remarkable high percentage of untreated CVD risk factors suggesting that not all opportunities to reduce the CVD burden are utilized.
INTRODUCTION
Healthy ageing is one of the topics in ‘Horizon 2020 – Personalising Health and Care’; “the biggest European Union Research and Innovation programme” aimed to ensure Europe’s global competitiveness1. The goal of Horizon 2020 is to gain insight in factors
and interactions comprising the development and maintenance of good health and the presence and progression of common diseases and disabilities. Throughout life, underlying genetic make-up and modifiable lifestyle factors such as behaviour, environment and nutrition interact in this process in varying degrees.
Despite recent progress with novel therapies, a major threat to healthy ageing is cardiovascular disease (CVD)2-5. CVD affects the majority of adults over 60 years of age.
In 2012, it was estimated to be the cause of 17.3 million deaths worldwide6. In the EU,
the main cause of death is CVD and accounts for 1.9 million deaths every year2. CVD
also causes substantial morbidity with an annual hospital discharge rate of 2,400 per 100,000 population.
Epidemiologic studies in the past, including the Framingham Heart study initiated in 1948, have contributed enormously to our understanding of CVD and its risk factors7.
However, after identification of risk factors with large effect size the power of many previous studies to test for smaller effect sizes or gene-environment interactions is limited. In addition, these cohorts date back to the 90s, and advances in treatment as well as changes in behavior and lifestyle have occurred. To further our knowledge of genes, environment and their interaction determining CVD and healthy ageing, contemporary population-based biobanks are essential. The LifeLines cohort study, established in 2006, is a contemporary observational population-based study designed to enhance our understanding of healthy ageing in the 21st century8. Baseline characteristics of
167,729 inhabitants of the Northern part of the Netherlands have been collected. The first follow-up visit at five years is ongoing and the second 10-year follow-up visit is scheduled. LifeLines participants will be followed up to 30 years. LifeLines is a facility that is open for all researchers, information on application and data access procedures is summarized on www.lifelines.net. Here we summarize the baseline characteristics, and provide detailed information on the prevalence of CVD, cardiovascular risk factors and treatment thereof. In addition, we aim to inform and encourage researchers to consider LifeLines cohort study for their future research projects.
METHODS
Overall Design of the LifeLines cohort study
The overall design and rationale of the LifeLines cohort study have been described in detail elsewhere8,9. In brief, individuals living in the recruitment area aged between
25 and 50, were invited through their general practitioners (GP). Individuals were not invited when the participating GP considered the patient not eligible by reason of severe psychiatric of physical illness; limited life expectancy or insufficient knowledge of the Dutch language. In addition, inhabitants of the Northern provinces, who were not invited by their GP and not meeting above-mentioned reasons, could register themselves via the LifeLines website. After signing informed consent, participants received a baseline questionnaire and an invitation to a health assessment at one of the LifeLines research sites. During these visits, participants were asked whether their family members would also be willing to participate. Overall, 49% of the participants (n=81,652) were invited through their GP, 38% (n=64,489) via participating family members and 13% (n=21,588) self-registered via the LifeLines website. In total, 167,729 participants were included from the end of 2006 until December 2013 and data of 167,016 participants were suitable for further analysis. The 5-year follow-up visit physical examination at the LifeLines research site is currently ongoing and the 10-year follow-up visit is planned. In addition, participants receive a follow-up questionnaire every 18 months. By using a third-party pseudo-anonymization system, records of GPs, pharmacies and other health and national registries are being linked with the LifeLines database. Data was analysed for different pre-specified age categories, namely juvenile (aged 8-18 years), young and middle-age adults (≥18 and <65) and older aged (65+) participants. Data collection within LifeLines is dynamic, add-on studies are continuously implemented in LifeLines.
Cardiovascular Data collection
Questionnaires
Self-reported questionnaires were used to obtain information on demographics, family composition, work and education, general health, lifestyle, environmental and psychosocial factors. Lifestyle and environment questions included information on physical activity (SQUASH questionnaire), nutrition (FFQ questionnaire), smoking, physical environment and daytime activities. Psychosocial factors included questions on perceived quality of life, health perception, personality, stress and social support8. Drug
use was collected in the questionnaire and categorized using the general Anatomical Therapeutic Chemical Classification System (ATC) codes. We recently reported a global overview of the definitions of CVD and non-CVD in a subpopulation of LifeLines10.
Physical examination
At baseline, participants were invited to visit one of twelve LifeLines research sites to undergo a physical examination and a series of tests. During the baseline visits height without shoes was measured with the SECA 222 stadiometer and rounded to the nearest 0.5 cm. Weight without shoes and heavy clothing was measured with SECA 761 scale and rounded to the nearest 0.1 kg. Waist and hip circumference were measured with SECA 200 measuring tape and rounded to the nearest 0.5 cm. Blood pressure was measured ten times during ten minutes with Dynamap, PRO 100V2. The blood pressure registered was calculated by averaging the final three readings in mmHg. Heart rate was collected and reported in beats per minute. Pulmonary function was measured once with Welch Allyn version 1.6.0.489 and a 12-lead electrocardiogram (ECG) was recorded with a Welch Allyn DT100 machine. Skin autofluorescence was measured at the lower arm with advanced glycation end products (AGE)-reader (AGEreader, DiagnOptics Technologies B.V., The Netherlands).
Biomaterial collection and biobanking
At the research sites, blood and 24-hour urine was collected from participants and transported to the central LifeLines laboratory in Groningen. For performing clinical chemistry analyses on fresh blood and 24-hour urine samples, part of the samples was directly transferred to the central laboratory of the University Medical Centre Groningen (UMCG). From the remaining blood samples, part has been used for DNA isolation (from whole blood of all LifeLines participants aged 8 years and older) and was stored at -80°C. Normalized DNA was stored at 4°C. The remaining blood and 24-hour urine samples were stored at -80°C and are available for future research questions. In addition to blood and urine, faeces of more than 50,000 participants have been collected and a hair scalp will be collected from all participants during the first follow-up visit.
Genotyping data
Currently, genome-wide genotyping data is available of 13,436 participants. These data have been generated using the Illumina CytoSNP-12v2 array, after which they were called in GenomeStudio (Illumina, Inc., San Diego, California, USA). Quality control was performed with PLINK, after which 268,407 SNPs and 13,436 samples remained.
Ultra-low-dose CT imaging
A substudy (IMA-LIFE) is currently being established on ultra-low-dose CT scanning of the thorax. To determine normal values of lung density, bronchial wall thickness and coronary calcium by age and gender, 12,000 randomly assigned participants will undergo CT scanning after signing additional informed consent.
For a complete overview of the available LifeLines data visit the LifeLines website at www.lifelines.net and the online data catalogue at https://catalogue.lifelines.nl/.
Definitions
Cardiovascular risk factors
Self-reported CVD risk factors were defined as present when they were affirmatively answered in the questionnaire and as being absent when answered negatively or missing. In addition, physical examination data at baseline visit was used to define and validate CVD risk factors specified by the following criteria (Supplementary Figures 1-7) show the operationalization methods for defining cardiovascular risk factors). Overweight was defined as a body mass index (BMI) above 25.0kg/m2. In juvenile
participants, overweight was defined according to World Health Organization (WHO) child growth standards with BMI-for-age11.
Smoking included past and current smokers. Active smoking in adults was defined as having smoked the past month or now. Former adult smokers were defined as answering the question ‘have you stopped smoking’ confirmatively. Data on smoking was available in juveniles aged 13 years and over. Active smoking in juveniles was defined as answering the question “does your child still smoke” confirmatively. Former smoking was defined as answering the question “did your child smoke daily” confirmatively and followed by the question “does your child still smoke” answered negatively. The question “Being active for at least half an hour a day”, was the definition for active lifestyle in adults, which was obtained from the questionnaire as well. In juveniles aged 8 years and over active lifestyle was defined as doing sports or playing outside for more than 7 hours a week. Cancer and blood clotting disorders were considered to be present when they were affirmatively answered in the questionnaire. The Systematic Coronary Risk Evaluation Project (SCORE) risk was determined in adult participants with available cholesterol and blood pressure measurements12.
Cardiovascular disease
By questionnaire, participants were asked to report presence of CVD and related symptoms. Operationalization methods were generated for defining (silent) myocardial infarction (MI), heart failure and atrial fibrillation (Supplementary Figures 8-11). With the help of these operationalization methods self-reported CVD or related symptoms were validated with biomarkers or cardiovascular medication. The total number of CVD per participant was determined. The definition for CVD was based on the ICD-10 and included all CVD that could be verified in the LifeLines database; MI, heart failure,
atrial fibrillation, heart valve disorders, arrhythmia, aneurysm, stroke, thrombosis, atherosclerosis, narrowing carotid arteries and a history of coronary artery bypass grafting (CABG)13.
Statistical analysis
Normally distributed continuous variables were presented with means and standard deviations. Continuous variables not normally distributed were presented as medians with interquartile ranges (IQR) and categorical variables as percentages. The Chi-square test was used to compare frequencies of events in the middle (aged ≥18 and <65) and older (aged 65+) aged group. Differences in continuous variables, not normally distributed, were ascertained by two-sample Wilcoxon rank-sum (Mann-Whitney) test. Age and sex standardized estimates were calculated with standardized rates for the variables, defined as the weighted average of stratum-specific rates. These rates are averaged across the weights of the general population, based on the population distribution of age and sex of adults 18 years and over (13,060,511) in the Netherlands in 2010. This is implemented with the dstdize command in STATA, an algebraically equivalent of the Cochran’s formula14. Logistic regression was performed to assess the
correlation between cardiovascular risk factors and CVD, presented with odds ratios. Adjustments for family relations were performed with the cluster option. Analyses were performed with STATA/IC version 13.0 (StataCorp LP, College Station, Texas, USA).
Ethics policy
Declaration of Helsinki: The LifeLines cohort study state that the study complies with the Declaration of Helsinki. The local ethics committee approved the research protocol and informed consent was signed by every participant.
RESULTS
The LifeLines cohort study population included 68,850 male and 83,330 female adults and 7,231 male and 7,605 female juveniles. In total 60,401 participants were part of 42,351 families, including first-, second- and third-degree relatives, generating 112,050 family clusters. The age and gender distribution of LifeLines participants differed substantially from the general population distribution in the Netherlands (Figure 1 and Table 1).
Table 1.
D
emog
raphics and car
dio
vascular r
isk fac
tors in the Lif
eLines c ohor t study Char ac teristics Juv enile 8-18 y ears N A dults 18-65 y ears N A dults 65+ y ears N Cr ude estima te Standar diz ed estima te Age (y ears , mean ± SD ) 9.11 ± 4.7 14,836 43 ± 11 141,327 71 ± 5 10,853 -Female 51.3% 7,605 59.0% 83,330 52.7% 5,720 -Ethnicit y W hit e/East and W est E ur opean -97.9% 109,574 99.0% 7,484 -M edit er ranean or A rabic -0.4% 406 0.1% 9 -Black -0.2% 189 <0.1% 3 -A sian -0.5% 570 0.2% 17 -O ther -1.0% 1,136 0.6% 47 -Car dio vascular r isk fac tors Self-r epor ted h yper tension -19.9% 28,059 36.8% 3,993 21.1% 22.2% H yper tension 0.4% 57 22.5% 31,748 69.0% 7,487 25.8% 32.6% Self-r epor ted h yper cholest er olemia -11.5% 16,234 29.2% 3,168 12.8% 14.7% H yper cholest er olemia 0.2% 35 12.8% 18,102 43.8% 4,753 15.0% 19.1% Self-r epor ted diabet es mellitus -2.0% 2,819 9.8% 1,063 2.6% 3.5% Diabet es mellitus 0.2% 31 2.6% 3,646 11.7% 1,270 3.2% 4.4% Self-r epor ted k idney disease -0.5% 655 1.0% 103 0.5% 0.6% Kidney disease 0.3% 41 1.4% 1,910 11.8% 1,281 2.1% 4.6% O ver w eigh t 13.0% 1,933 54.0% 76,282 71.2% 7,730 55.2% 56.2% Ac tiv e smoker 1.2% 184 21.5% 30,412 8.2% 893 20.6% 19.0% For mer smoker 0.4% 65 32.5% 45,953 52.4% 5,687 33.9% 35.3% Ac tiv e lif est yle (30 min/da y) 42.4% 6,292 21.4% 30,257 23.7% 2,567 21.6% 21.1% Family health - C VD -8.9% 12,510 10.0% 1,083 8.9% 8.8% Canc er -3.8% 5,331 15.8% 1,709 4.6% 6.2% COPD -9.9% 14,037 23.4% 2,540 10.9% 12.6% Rheuma toid ar thr itis -1.2% 1,658 1.8% 199 1.2% 1.2% Th yr oid disease 0.2% 30 2.2% 3,074 4.9% 536 2.4% 2.6%
Blood clotting disor
der -0.6% 852 0.7% 77 0.6% 0.6% CVD = car dio vascular disease , C OPD = Chr onic Obstruc tiv e P ulmonar y Disease , N = number , SD = standar d devia tion
Figure 1. A. Population distribution in the Netherlands in 2010. B. Population distribution in the LifeLines cohort study, inclusion 2006-2013.
Baseline characteristics and cardiovascular risk factors
Physical examination was available on all participants. Lower weight was seen in the older aged group compared to the middle age category; in women the mean weight was 74±14 (old age category) and 74±12kg (middle age category, p=0.001) respectively and in men 88±14 and 85±11kg (p<0.001) respectively. The mean height of women in the middle age category was higher compared to the older age category (170±7cm versus 164±6cm, p<0.001). The height of men was lower in the older aged group: 183±7cm in the middle age category compared to 177±cm in the older aged group (p<0.001). In total 61.9% (n=35,878) of men were overweighed in the middle age category compared to 48.5% (n=40,450) of women (Figure 2). In the older age group, these proportions were higher: 73.5% (n=3,774, p<0.001) of men and 69.4% (n=3,967, p<0.001) of women. In contrast to the lower heart rate (72±11bpm in the young and middle-age adults versus 68±14bpm in the older aged, p<0.001), systolic blood pressure was higher in older age categories (124±15 over 74±9mmHg in the young and middle-age adults vs. 137±24 over 74±12mmHg in the older aged, p<0.001). Additional biomarkers and cardiovascular drug use were available (Supplementary Table 1 and Table 2).
The prevalence of cardiovascular risk factors is presented in Table 1. In 110,502 (72.6%) participants, at least one classical cardiovascular risk factors (hypertension, hypercholesterolemia, diabetes mellitus, kidney disease, overweight or current smoking) was present. As expected, the burden of cardiovascular risk factors (SCORE) increased with age (Figure 3A).
Cardiovascular drug use increased with age. Amongst the most prescribed cardiovascular drugs were agents acting on the renin-angiotensin system, lipid drugs and beta-blockers. Overall, 12.3% (n=17,341) of the participants in the young and middle-age adults used one or more cardiovascular drug compared to 54.8% (n=5,945, p<0.001) of the participants in the older aged group. However, of the participants with hypertension 75.2% had no antihypertensive drugs, for hypercholesterolemia 95.9% had no lipid lowering drugs, for diabetes 41.2% had no anti-diabetic drugs.
The recent SPRINT trial proclaimed lower rates of cardiovascular events in persons with strict blood pressure control (<120 mmHg)15. In a recent publication investigating
eligibility of SPRINT criteria in U.S.A. adults, more than half of the adults with hypertension were not treated16. In this cohort, of the participants with systolic blood pressure higher
than 130 mmHg (n=54,026, 35.5%) 79.9% (n=43,153) had no antihypertensive drugs. Of participants having a SCORE predicting a ≥5% 10-year risk of fatal CVD, 53.2% were not using any cardiovascular preventive drugs.
Body Mass Index
BMI (kg/m2) indi vi dua ls (% ) <18.5 18.5-25.0 25.0-30.0 30.0-35.0 35.0-40.0 40.0 0 10 20 30 40 50 60 Men 65+ yrs Women 18-65 yrs Women 65+ yrs Men 18-65 yrs
Figure 2. Body Mass Index distribution for males and females in the middle and older age category in the LifeLines cohort study, BMI = Body Mass Index
Cardiovascular disease
Symptoms possibly related to CVD were common. Chest pain and dyspnea on exertion were reported by approximately 25% of all adult participants. CVD and suggestive symptoms along are presented in Table 2. CVD was present in 16,872 (11%) of adult participants and increased with age from 9% in the middle age category to 28% in the older age category. Figure 3B shows total CVD burden in the LifeLines cohort study population. Prevalence of CVD increases substantially from the age of 50 (p<0.001). Interestingly, we observed that not only the number of individuals with CVD increases, but the amount of reported CVD manifestations increases as well. Table 3 shows multivariate logistic regression of CVD risk factors with corresponding odds ratios for CVD. After adjustment for family clusters odds ratios were analogous.
SCORE in LifeLines age (years) SC O R E (% ) 18-29 30-39 40-49 50-59 60-69 70-79 80+ 1.0 2.0 3.0 4.0 CVD in LifeLines age (years) indi vi dua ls (% ) 18-29 30-39 40-49 50-59 60-69 70-79 80+ 10 20 30 40 50 1 CVD 2 CVD 3 CVD 4 CVD SCORE in LifeLines age (years) SC O R E (% ) 18-29 30-39 40-49 50-59 60-69 70-79 80+ 1.0 2.0 3.0 4.0 CVD in LifeLines age (years) indi vi dua ls (% ) 18-29 30-39 40-49 50-59 60-69 70-79 80+ 10 20 30 40 50 1 CVD 2 CVD 3 CVD 4 CVD
Figure 3. A. Mean SCORE in LifeLines per age category. B. Mean number of CVD manifestations in the LifeLines cohort study. CVD = Cardiovascular disease, SCORE = Systematic Coronary Risk Evaluation Project
Table 2.
C
ar
dio
vascular disease in the Lif
eLines c ohor t study Char ac teristics A dults 18-65Y N A dults 65+Y N Cr ude estima te Standar diz ed estima te MI, hear t failur e and a trial fibrilla tion Self-r epor ted MI 0.7% 985 6.2% 665 1.1% 2.0% W
ith drug use or EC
G abnor malities 0.6% 852 5.6% 608 1.0% 1.8% Silen t MI 0.1% 162 0.5% 58 0.1% 0.2% Possible diag nosis silen t MI 0.4% 549 1.4% 152 0.5% 0.7% Self-r epor ted hear t failur e 0.6% 776 3.2% 342 0.7% 1.2% W
ith drug use or ther
ap y other wise 0.4% 495 2.8% 307 0.5% 1.0% Car diac implan table elec tr onic devic e 0.1% 139 0.8% 87 0.2% 0.3% Tr ansplan t <0.1% 10 <0.1% 14 <0.1% <0.1% A tr ial fibr illa tion 0.3% 408 3.9% 426 0.6% 1.3% O ther self -r ep or ted C VD Balloon ang ioplast y or b ypass sur ger y 0.9% 1,323 8.4% 898 1.5% 2.8% Hear t v alv e disor der 0.9% 1,237 3.1% 332 1.0% 1.4% Palpita tions 6.7% 9,462 14.6% 1,579 7.3% 8.0% Aor tic aneur ysm 0.2% 266 1.9% 202 0.3% 0.6% Str oke 0.6% 842 3.1% 336 0.8% 1.2% Thr ombosis 1.1% 1,550 2.9% 316 1.2% 1.4% A ther oscler osis 0.4% 491 2.0% 214 0.5% 0.8% Nar ro wing car otid ar ter ies 0.2% 271 1.5% 124 0.3% 0.4% Sympt oms Per ipher al edema 14.4% 20,038 23.9% 2,082 14.5% 13.9% Chest pain 26.5% 36,972 28.1% 2,446 25.9% 25.8% Shor tness of br ea th 22.9% 31,567 19.6% 1,703 21.9% 20.8% D yspnea 10.3% 14,436 6.6% 572 9.9% 9.1% D yspnea on e xer tion 25.9% 36,087 27.7% 2,403 25.3% 24.7% Or thopnea 3.5% 4,899 4.2% 360 3.5% 3.2% CVD = car dio vascular disease , EC G = elec tr ocar diog ram, MI = m yocar dial infar ction
Table 3. Multivariate logistic regression: risk factors and CVD
Characteristics p-value Odds ratio 95% CI
Female <0.001 1.08 1.04-1.12
Age per year <0.001 1.04 1.03-1.04
Overweight 0.003 1.06 1.02-1.10
Smoking <0.001 1.21 1.17-1.26
Active lifestyle
Active 0 out of 7 days Reference - -Active 1 out of 7 days <0.001 0.82 0.74-0.90 Active 2 out of 7 days <0.001 0.80 0.73-0.88 Active 3 out of 7 days <0.001 0.80 0.73-0.88 Active 4 out of 7 days <0.001 0.78 0.72-0.86 Active 5 out of 7 days <0.001 0.84 0.77-0.92 Active 6 out of 7 days <0.001 0.84 0.77-0.92 Active 7 out of 7 days <0.001 0.86 0.79-0.93
Family CVD <0.001 1.30 1.24-1.38 Cancer <0.001 1.15 1.08-1.24 Thyroid disease <0.001 1.53 1.39-1.68 Kidney disease <0.001 1.23 1.11-1.36 Hypercholesterolemia <0.001 1.84 1.77-1.92 Hypertension <0.001 2.14 2.06-2.23 Diabetes 0.025 1.10 1.01-1.19
CI = confidence interval, CVD = cardiovascular disease
DISCUSSION
Here we describe the baseline cardiovascular characteristics of the contemporary three-generations LifeLines cohort study with 167,016 participants. The risk factor burden of the LifeLines cohort study is high and can be extrapolated to the Dutch general population by adjusting for population distribution. Over 70% of the participants had at least one cardiovascular risk factor (hypertension, hypercholesterolemia, diabetes mellitus, kidney disease, overweight or current smoking) and in a substantial proportion (11%) a manifestation of CVD was present. Primary prevention of cardiovascular risk factors, even when SCORE predicted a 5% or more risk, was remarkably low. The burden of risk factors and CVD present in the LifeLines cohort study provides considerable
power to study events and risk factors related to CVD. It is therefore a valuable tool for researchers to further study the role of CVD and its risk factors in relation to healthy ageing.
The LifeLines study population differs from the general population by its design to that effect that the proportion of adults aged 25-50 years are overrepresented17. Reported
prevalences of CVD risk factors from the Dutch National Registry (Statistics Netherlands) based on national health survey in around 15,000 persons, are frequently lower compared to the LifeLines cohort study17. This may be due to different methods used for
identifying and defining disease. For example, according to the Statistics Netherlands for 2013 and 2014 prevalence of hypertension and overweight were around 11% lower17. Smoking was estimated 4% higher than in the LifeLines cohort study, with a
prevalence of 24.9% compared to 20.6%. In contrast, the WHO reported generally higher prevalences with hypercholesterolemia, diabetes and overweight estimated 5% higher than in the LifeLines study18.
Discrepancies exist regarding physical activity. According to the WHO in 2010 17.9% of adults from the Netherlands were insufficiently active18 and the Statistics Netherlands
reported 63% of the population attained sufficient physical activity. These differences might be due to the use of different definitions and measurements of physical activity. The reported family history of CVD was four times higher in the Rotterdam study compared to the LifeLines, suggesting regional differences, the use of different definitions, or underreporting in the LifeLines cohort study19. MI and stroke percentages in the
LifeLines cohort study were somewhat lower compared to the Statistics Netherlands inquiry in 2014, in which 3.1% of the total population ever had a stroke and 3.3% had a MI compared to respectively 0.8% and 1.1% in the LifeLines study population.
In the Netherlands in 2012, drug use of lipid lowering drugs was 10.7%, beta-blockers 9.8% and diabetes 4.6%, similar as reported by the participants of the LifeLines cohort study17. Interestingly, taking into account the risk factor burden in participants with a
predicted risk of ≥5% of which the majority does not use cardiovascular preventative drugs, there is likely to be a considerable underutilisation of primary prevention. The general practitioner has been informed about the risk profile of the LifeLines participant as part of the protocol and in line with the recommendation of the Medical Ethical Committee. Follow-up studies will be performed to study whether this knowledge has increased the percentage of subjects in whom primary prevention was initiated. Strengths of the Lifelines cohort study include the open protocol and hence a continuous possibility for researches to implement add-on studies. The three-generation structure
in combination with the available genome-wide genetic data enables unique opportunities for the analysis of genetic traits. In LifeLines, over one-thirds of the participants had first-, second-, or third-degree relatives also taking part in the study. The family design of the LifeLines cohort study has advantages with respect to multiple-level information, separation of non-genetic and genetic familial transmission and the investigation of (epi)genetic influences. Another advantage when performing genetic research, is the relatively homogeneous study population due to a low migration rate in the northern part of the Netherlands (net migration rate of 0.80 per 1,000 inhabitants in 2012)17. Less than 2% of the total included population had an ethnicity other than
white-, east- or west-European. Diversity regarding CVD exists within ethnicity groups, and reported prevalences were not corrected for ethnicity. Current data might not be applicable to other ethnicities.
CONCLUSION
The LifeLines cohort study is a large population based cohort accessible to national and international researchers8. The three-generation structure in combination with the
available genome-wide genetic data enables unique opportunities for the analysis of environmental and genetic traits. The family design of LifeLines enables inclusions of three generation families and has advantages with respect to multiple-level information, separation of non-genetic and genetic familial transmission. The prevalence of CVD risk factors and conditions is abundant in LifeLines and enables researchers to improve our knowledge on CVD and healthy cardiovascular ageing. A remarkable high percentage of untreated CVD risk factors in LifeLines suggest that not all opportunities to reduce the CVD burden are utilised.
Acknowledgements
The authors wish to acknowledge the services of the LifeLines cohort study, the contributing research centres delivering data to LifeLines, and all the study participants.
Funding
The LifeLines cohort study, and generation and management of GWAS genotype data for the LifeLines cohort study is supported by the Netherlands Organization of Scientific Research NWO (grant 175.010.2007.006), the Economic Structure Enhancing Fund (FES) of the Dutch government, the Ministry of Economic Affairs, the Ministry of Education, Culture and Science, the Ministry for Health, Welfare and Sports, the Northern
Netherlands Collaboration of Provinces (SNN), the Province of Groningen, University Medical Center Groningen, the University of Groningen, Dutch Kidney Foundation and Dutch Diabetes Research Foundation.
Disclosures
REFERENCES
1. Horizon 2020. Work programme 2014-2015. 2014:7.
2. European Heart Network and European Society of Cardiology. European cardiovascular disease statistics 2012 edition. 2012;4th edition.
3. Vermond RA, Geelhoed B, Verweij N, et al. Incidence of atrial fibrillation and relationship with cardiovascular events, heart failure, and mortality: A community-based study from the netherlands. J Am Coll Cardiol. 2015;66(9):1000-1007.
4. Nichols M, Townsend N, Scarborough P, Rayner M. Cardiovascular disease in europe 2014: Epidemiological update. Eur Heart J. 2014;35(42):2929.
5. Sipido KR, Van de Werf F. The growing burden of cardiovascular disease. Eur Heart J. 2012;33(13):1540-1541.
6. Laslett LJ, Alagona P,Jr, Clark BA,3rd, et al. The worldwide environment of cardiovascular disease: Prevalence, diagnosis, therapy, and policy issues: A report from the american college of cardiology. J Am Coll Cardiol. 2012;60(25 Suppl):S1-49.
7. Fuster V, Kelly B. Promoting cardiovascular health in the developing world: A critical challenge
to achieve global health. Washington, D.C.: The National Academies Press; 2010.
8. Scholtens S, Smidt N, Swertz MA, et al. Cohort profile: LifeLines, a three-generation cohort
study and biobank. Int J Epidemiol. 2014.
9. Stolk RP, Rosmalen JG, Postma DS, et al. Universal risk factors for multifactorial diseases: LifeLines: A three-generation population-based study. Eur J Epidemiol. 2008;23(1):67-74. 10. Meems LM, de Borst MH, Postma DS, et al. Low levels of vitamin D are associated with
multimorbidity: Results from the LifeLines cohort study. Ann Med. 2015;47(6):474-481. 11. World Health Organization. Child growth standards. http://www.who.int/childgrowth/
standards/bmi_for_age/en/. Updated 2015. Accessed August, 1st, 2015.
12. Perk J, De Backer G, Gohlke H, et al. European guidelines on cardiovascular disease prevention in clinical practice (version 2012) : The fifth joint task force of the european society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of nine societies and by invited experts). Int J Behav
Med. 2012;19(4):403-488.
13. World Health Organization. International classification of diseases (ICD). http://www.who. int/classifications/icd/en/. Updated 2015. Accessed 08, 2015.
14. Cochran WG. Sampling techniques, 3rd edition. 3rd ed. ; 1977.
15. SPRINT Research Group. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med. 2015.
16. Bress AP, Tanner RM, Hess R, Colantonio LD, Shimbo D, Muntner P. Generalizability of results from the systolic blood pressure intervention trial (SPRINT) to the US adult population. J Am
Coll Cardiol. 2015.
17. Statistics Netherlands. StatLine databank. http://statline.cbs.nl/Statweb/?LA=nl. Updated 2015. Accessed July, 14th., 2015.
18. World Health Organization. Global health observatory. http://www.who.int/gho/countries/ en/. Updated 2015. Accessed July, 14th., 2015.
19. van Dis I, Kromhout D, Boer JM, Geleijnse JM, Verschuren WM. Paternal and maternal history of myocardial infarction and cardiovascular diseases incidence in a dutch cohort of middle-aged persons. PLoS One. 2011;6(12):e28697.
The effect of feedback on cardiovascular risk
factors on optimization of primary prevention:
a population based cohort study
• • •
M. Yldau van der Ende, Ingmar E. Waardenburg, Erik Lipsic, Harold Snieder, Pim van der Harst
ABSTRACT
Background
Primary prevention of cardiovascular disease (CVD) is an important strategy to further reduce the number of cardiovascular events and associated health care costs. It is unknown whether population based single assessment of CVD risk and feedback to individuals and primary care physicians results in initiation of preventive cardiovascular pharmacotherapy in those at risk.
Methods
At the baseline visit of the Lifelines cohort study information on cardiovascular risk factors was collected of all participants. Abnormal lipid levels and abnormal blood pressure was reported to the participants as well as to their general practitioners. Pharmacotherapy could be obtained by linking the Lifelines participants to the IADB.nl pharmacy database (N=48,770). An interrupted time series analysis was plotted, in which the start year of blood pressure and lipid lowering medication was displayed in years before or after the baseline visit. Subsequently, predictors of the initiation of pharmacotherapy were determined and possible reduction in cardiovascular events that could be achieved by optimal treatment of individuals at risk.
Results
Before the Lifelines baseline visit, 34% (out of 1,527, 95% Confidence interval (CI) 32%-36%) and 30% (out of 1,991, 95% CI 28%-32%) of the individuals at risk had a blood pressure or lipid lowering drug prescription, respectively. In individuals at risk, the use of blood pressure lowering medication, but not lipid lowering medication, increased substantially during the year of the baseline visit. Treating all individuals at increased risk, but without preventive medication (≥5% 10-year risk) with lipid or blood pressure lowering medication (N=8,515 and N=6,899) would have prevented 162 and 183 CVD events, respectively, in the upcoming five years.
Conclusion
Primary prevention of CVD in the general population appears suboptimal. A single visit and feedback of cardiovascular risk factors, including lipids levels and blood pressure, of participants of the Lifelines cohort study resulted in a substantial increase of blood pressure lowering medication and extrapolated health benefits.
INTRODUCTION
In the European Union, cardiovascular death accounts for approximately 37% of total mortality rates1. It is thought that a substantial part of cardiovascular mortality can be
reduced by early identification and treatment of cardiovascular risk factors1,2. Despite
numerous studies underlining the importance of early detection3,4, contradictory
evidence is reported on the efficacy of population screening5,6. The number of
individuals with cardiovascular risk factors is, however, expecting to rise till 20307,
driving governments to focus on healthy aging, including cardiovascular disease (CVD) prevention. In the Netherlands, general practitioners (GP) deliver first-line healthcare and have a major role in identification and treatment of individuals at risk for CVD. Identification is usually based on case finding. During a regular visit, the GP decides whether or not to further investigate the presence of cardiovascular risk factors by making inquiries about lifestyle habits and by measuring blood pressure and serum lipid levels. This results in a cardiovascular risk profile, based on the Systematic Coronary Risk Evaluation (SCORE)8 and a recommendation regarding the start of preventive medication
use. Although case finding can be improved via population based studies it remains to be determined whether providing feedback of risk factors via population based studies to participants and GPs results in the initiation of preventive pharmacotherapy. The Lifelines cohort study collects data from 167,729 individuals of the Northern part of the Netherlands9,10. During the baseline visit, data on cardiovascular risk factors, including
blood pressure and blood lipid levels, is collected and reported to the participants and their GPs. The aim of the current study is to determine the effect of providing feedback on risk factors via a population based study, on initiation of preventive pharmacotherapy.
METHODS
Study design and subjects of the Lifelines cohort study
The Lifelines cohort study is a multi-disciplinary prospective population-based cohort study examining in a unique three-generation design the health and health-related behaviors of 167,729 persons living in the North of The Netherlands. It comprises a broad range of investigative procedures in assessing the biomedical, socio-demographic, behavioral, physical and psychological factors which contribute to the health and disease of the general population, with a special focus on multi-morbidity and complex genetics. The study design and rationale of Lifelines were previously described in detail9. During the baseline visit an informed consent form was signed, and blood and
