Abstract

Background

Primary prevention of cardiovascular disease (CVD) relies on the identification of individuals at increased risk of developing cardiovascular events. Circulating biomarkers mirroring the (subclinical) disease process are valuable tools for CVD risk prediction. Evidence is accumulating that the clinical presentation and mechanisms for CVD differ between men and women. A systematic review of sex-specific data was performed on biomarker levels and their association with CVD in primary prevention in order to investigate the availability of sex-specific data and to explore for any differences in the associations between men and women.

Methods and results

PubMed MEDLINE and Embase were searched on 2 February 2014 and updated on 15 January 2015. Biomarkers included represented pathophysiological pathways of lipids, inflammation, kidney function, and of the heart. Data on patient characteristics, sex-specific biomarker levels, biomarker association with future CVD events and clinical value were extracted. Only 54 studies of 360 publications provided sex-specific information. Most of the remaining 306 publications not providing sex-specific results only corrected for sex in multivariable models. The additional clinical utility of biomarkers was reported in seven publications, one of which was stratified by sex.

Conclusion

Sex-specific data on biomarkers for CVD in the general population exist, but it is underreported. There is inconsistency in sex-specific differences in levels of traditional biomarkers and in their relation to CVD. To improve personalized cardiovascular diagnoses and care for men and women, reporting sex-specific data on clinical utility of biomarkers is crucial and should be encouraged in publications of sufficiently powered studies.

See page 64 for the editorial comment on this article (doi:10.1093/ehjqcco/qcw006)

Introduction

Cardiovascular disease (CVD) remains the leading cause of death and disability worldwide. Women tend to develop CVD ∼10 years later than men. However, the global scale of the problem for women is evident and should not be underappreciated; heart disease is the leading cause of death in women in every major developed country and in most emerging economies.1 Early and accurate CVD risk prediction and the subsequent implementation of prevention strategies play vital roles towards combating CVD in both men and women. Identifying those at risk of heart disease prior to the onset not only improves the health of the population but will also reduce long-term healthcare expenditure as CVD is the leading disease for direct and indirect healthcare expenditure, accounting for 320 billion dollars in the USA alone in 2011.2 Several algorithms have been developed to facilitate risk prediction in individual patients. Most include sex as a risk factor alongside traditional risk factors such as blood pressure, smoking, diabetes, and lipid values.3 Considerable effort has been placed upon improving cardiovascular risk prediction with the use of biomarkers that play a role in the pathogenesis of the disease or reflect atherosclerosis status. It is becoming increasingly evident that the disease mechanisms of CVD in women differ compared with men. For example, autopsy studies reveal that women with an acute myocardial infarction are more likely to have plaque erosion as a substrate for the event whereas in men plaque ruptures are more likely to precipitate the event.4 Atherosclerotic carotid plaque composition also differs between men and women undergoing carotid endarterectomy independent of presenting symptoms.5,6 Cardiac disease presentation and mechanisms may also vary between sexes. Systolic heart failure with reduced ejection fraction is more prevalent in males while diastolic heart failure with preserved ejection fraction is strikingly more prevalent in women than men with a ratio of 2:1.7 In addition, women presenting with stable cardiovascular symptoms undergoing coronary angiogram have less severe coronary artery disease as measured by SYNTAX score.8 Sex differences in the magnitude of risk factors with CVD have been reported.9,10 Given these notable differences, the American Heart Association has issued female-specific clinical guidelines for CVD prevention since 1999. The most recent update in 2011 focused on recommendations that are effective in clinical practice.11 There are several female-specific factors such as menopause and preeclampsia which in the past have been suggested to be related to the pathophysiology of CVD. Currently, there remains no clear evidence as to whether these factors increase the risk of CVD in women or not. Differences in traditional risk factors such as smoking between men and women have also been reported with a prolonged history of smoking found to be more hazardous in women than in men.12 Despite these discrepancies, data on potential sex differences in established and emerging biomarkers for CVD prediction in the general population are lacking. Therefore, we undertook a systematic review to investigate the availability of sex-specific evidence of plasma-based biomarker levels and their association with CVD in primary prevention cohorts and subsequently examined the sex-specific associations found. For this review, we selected biomarkers that are either incorporated in current CVD risk prediction scores or that are emerging in the field.

Methods

Eligibility criteria and selection of studies

A systematic search was performed at PubMed MEDLINE (http://www.ncbi.nlm.nih.gov) and Embase (www.embase.com) on 14 February 2014 using the strings described in Supplementary material online, Table S1. The search was repeated on 15 January 2015 only including papers published from 14 February 2014 onwards. Duplicate papers found both from Medline and Embase were removed. Inclusion criteria for the selection of papers included the correct biomarker, the correct outcome and the correct domain. Conference abstracts and review papers were removed. Only full-text papers were included. Biomarkers that were eligible for inclusion were those that are currently being used in primary prevention or those that are emerging in the field. The latter were selected in consensus with the BiomarCaRE consortium (Biomarker for Cardiovascular Risk Assessment in Europe). BiomarCaRE is an EU FP7-funded collaborative research project that integrates experts from clinical, epidemiological, and biomarker research, as well as commercial enterprises throughout Europe.13 A list of the selected biomarkers is shown in Table 1. The outcomes of interest included in the search were cardiovascular endpoints coronary heart disease (CHD), heart failure, stroke, and atrial fibrillation (AF) measured in either cohort studies, in a case–cohort setting, or case–control setting of a cohort study. The domain consisted of individuals free from CVD at baseline. All fields were searched for any term related to gender or sex (see Supplementary material online,Table S1). These terms included female/females, woman/women, and the male equivalents. Sex/sexes, gender, gender-stratified, and gender-specific terms were also included in the search. There was no limit to the year of publication in the search criteria. Only papers in English language were included.

Table 1

A list of the selected biomarkers eligible for inclusion

Biomarkers of lipid metabolism:
  • LDL-cholesterol

  • HDL-cholesterol

  • Triglycerides

  • Lipoprotein(a)

  • ApoA1

  • ApoB2

 
Biomarkers of inflammation:
  • C-reactive protein

  • GDF-15

 
Biomarkers of kidney function:
  • Glomerular filtration rate

  • Creatinine

  • Cystatin C

 
Insulin and glucose 
Biomarkers of myocyte necrosis:
  • Troponin

 
Biomarkers of myocyte stress:
  • B-type natriuretic peptide

  • N-terminal pro B-type natriuretic peptide

  • Galectin-3

 
Others:
  • ST2/interleukin 1 receptor-like 1

  • Vitamin D

  • miRNAs

  • Metabolomics

 
 
Biomarkers of lipid metabolism:
  • LDL-cholesterol

  • HDL-cholesterol

  • Triglycerides

  • Lipoprotein(a)

  • ApoA1

  • ApoB2

 
Biomarkers of inflammation:
  • C-reactive protein

  • GDF-15

 
Biomarkers of kidney function:
  • Glomerular filtration rate

  • Creatinine

  • Cystatin C

 
Insulin and glucose 
Biomarkers of myocyte necrosis:
  • Troponin

 
Biomarkers of myocyte stress:
  • B-type natriuretic peptide

  • N-terminal pro B-type natriuretic peptide

  • Galectin-3

 
Others:
  • ST2/interleukin 1 receptor-like 1

  • Vitamin D

  • miRNAs

  • Metabolomics

 
 

Data extraction

Publications were reviewed in duplicate (Hester den Ruijter and Renate B. Schnabel). A third reviewer (Aisha Gohar) reviewed the additional publications from the repeated search. The following data were then selected: the study design, number of individuals included, mean age, baseline values of the biomarkers, outcome, median follow-up time, and the reported association with CVD. Associations of biomarkers with outcome were extracted as hazard ratios (HRs), odds ratios (ORs), risk ratios (RRs), or population attributable risks (PARs). When reported, data on the predictive performance in terms of discrimination and (re)classification were also obtained. Levels of biomarkers and the association with CVD were considered different between sexes when mentioned as such in the manuscript.

Risk of bias assessment

To assess the risk of bias in each of the 54 publications, the Nottingham-Ottawa scale was used,14 a tool recommended by the Cochrane Collaboration in the assessment of observational studies. This involves a star-rating system to grade each study on the basis of three domains in both case–control studies and cohort studies: selection of participants, comparability, and the ascertainment of exposure or outcomes of interest for case–control studies and cohort studies, respectively. The studies receiving the highest number of stars, ten for case–control studies, eight for cohort studies, and nine for case–cohort studies were deemed to have the lowest risk of bias. Studies with a star rating of eight or nine for case–control studies, six or seven stars for cohort studies, and seven or eight for case–cohort studies were deemed to have an intermediate risk of bias. Those with the lowest star rating were deemed to have the highest risk of bias.

Results

Figure 1 shows the flowchart of the selection process of publications. Our initial search (Table 2) resulted in 5374 articles. Following the removal of duplicates, reviewing titles, abstracts and full-text, 360 papers met our search criteria of the correct biomarker, outcome and domain. Of these 360 papers, 135 did not show any sex stratification (37.5%) and 171 papers only reported on one sex, either men or women (47.5%). Therefore, the number of remaining papers showing sex-stratified data regarding the chosen biomarkers and their association with CVD was 54 (15%). From the 54 articles included in this review, 46 (85%) articles reported sex-specific data on biomarkers and CVD in general population cohort studies and eight articles reported sex-specific data on biomarkers and CVD in case–control studies (n = 6) and case–cohort studies (n = 2). Full details of the 54 papers can be found in Supplementary material online, tables including the study design, baseline characteristics of the men and women including study numbers and mean ages, baseline values of the biomarkers, outcome, median follow-up time, and the reported association with CVD.

Table 2

Percentages of studies reporting differences in sex-specific association with CVD and biomarker levels

Biomarker Number of studies Total number of women (w)/men (m) included in the studies Number of studies reporting baseline sex differences Number of studies reporting baseline levels higher in women (w)/men (m) Sex-specific association with CVD CVD association stronger in women/men 
TC 161 834 (w), 92 625 (m) 3 (38%) 2 (w)/1 (m) 1 (13%) 1 in men 
LDL-C 64 399 (w), 190 399 (m) 1 (13%) 2 (25%) 1 in women, 1 in men 
HDL-C 13 364 593 (w), 29 067 (m) 5 (38%) 5 (w) 1 (8%) 1 in women 
Triglycerides 259 446 (w), 195 928 (m) 2 (22%) 2 (m) 2 (22%) 2 in women 
TC/HDL ratio 4112 (w), 2732 (m) 0 (0%) – 0 (0%) – 
Lp(A) 22 511 (w), 17 635 (m) 3 (50%) 3 (w) 4 (67%) 2 in women, 2 in men 
ApoA1 40 225 (w), 51 687 (m) 1 (25%) 1 (w) 0 (0%) – 
ApoB 40 225 (w), 51 687 (m) 1 (25%) 1 (m) 0 (0%) – 
CRP 5928 (w), 5879 (m) 3 (38%) 3 (m) 1 (25%) 1 in men (with two different outcomes) 
GFR 99 143 (w), 58 478 (m) 0 (0%) – 0 (0%) – 
Cystatin C 3517 (w), 3136 (m) 0 (0%) – 0 (0%) – 
Creatinine 64 185 (w), 67 800 (m) 0 (0%) – 1 (50%) 1 in women 
HbA1c 1498 (w), 1232 (m) 0 (0%) – 1 (100%) 1 in women 
Glucose 190 247 (w), 128 144 (m) 1 (33%) 1 (m) 0 (0%) – 
Insulin 968 (w), 541 (m) 0 (0%) – 1 (100%) 1 in men 
Troponin T 7687 (w), 5557 (m) 2 (100%) 2 (m) 0 (0%) – 
BNP 22 411 (w), 12 394 (m) 2 (50%) 3 (w) 1 (25%) 1 in men 
Vitamin D 5497 (w), 6456 (m) 1 (50%) – 2 (100%) 2 in women 
miRNAs 409 (w), 411 (m) 0 (0%) – 0 (0%) – 
ST2 4219 (w), 4225 (m) 1 (100%) 1 (m) 1 (100%) – 
Biomarker Number of studies Total number of women (w)/men (m) included in the studies Number of studies reporting baseline sex differences Number of studies reporting baseline levels higher in women (w)/men (m) Sex-specific association with CVD CVD association stronger in women/men 
TC 161 834 (w), 92 625 (m) 3 (38%) 2 (w)/1 (m) 1 (13%) 1 in men 
LDL-C 64 399 (w), 190 399 (m) 1 (13%) 2 (25%) 1 in women, 1 in men 
HDL-C 13 364 593 (w), 29 067 (m) 5 (38%) 5 (w) 1 (8%) 1 in women 
Triglycerides 259 446 (w), 195 928 (m) 2 (22%) 2 (m) 2 (22%) 2 in women 
TC/HDL ratio 4112 (w), 2732 (m) 0 (0%) – 0 (0%) – 
Lp(A) 22 511 (w), 17 635 (m) 3 (50%) 3 (w) 4 (67%) 2 in women, 2 in men 
ApoA1 40 225 (w), 51 687 (m) 1 (25%) 1 (w) 0 (0%) – 
ApoB 40 225 (w), 51 687 (m) 1 (25%) 1 (m) 0 (0%) – 
CRP 5928 (w), 5879 (m) 3 (38%) 3 (m) 1 (25%) 1 in men (with two different outcomes) 
GFR 99 143 (w), 58 478 (m) 0 (0%) – 0 (0%) – 
Cystatin C 3517 (w), 3136 (m) 0 (0%) – 0 (0%) – 
Creatinine 64 185 (w), 67 800 (m) 0 (0%) – 1 (50%) 1 in women 
HbA1c 1498 (w), 1232 (m) 0 (0%) – 1 (100%) 1 in women 
Glucose 190 247 (w), 128 144 (m) 1 (33%) 1 (m) 0 (0%) – 
Insulin 968 (w), 541 (m) 0 (0%) – 1 (100%) 1 in men 
Troponin T 7687 (w), 5557 (m) 2 (100%) 2 (m) 0 (0%) – 
BNP 22 411 (w), 12 394 (m) 2 (50%) 3 (w) 1 (25%) 1 in men 
Vitamin D 5497 (w), 6456 (m) 1 (50%) – 2 (100%) 2 in women 
miRNAs 409 (w), 411 (m) 0 (0%) – 0 (0%) – 
ST2 4219 (w), 4225 (m) 1 (100%) 1 (m) 1 (100%) – 
Figure 1

Flow chart showing the selection process of the systematic review with resulting publications.

Figure 1

Flow chart showing the selection process of the systematic review with resulting publications.

Results for each biomarker category are summarized below.

Biomarkers of lipid metabolism

The majority of the papers that provided sex-stratified biomarker data were regarding lipids. Supplementary material online, Tables S2–S7 shows the results reported in the publications for the selected lipid biomarkers. Sex-specific data on total cholesterol, LDL-C, HDL-C, and triglycerides were available in 8, 8, 13, and 9 papers, respectively. One paper also reported on the total cholesterol/HDL ratio. Most of the papers reported on the endpoint CHD (see Supplementary material online, Table S2). Two studies reported higher baseline total cholesterol levels in women when compared with men,15,16 with no differences in the association with CHD. Only one study reported higher baseline levels in men when compared with women,17 also with no differences in the association with CHD. All reported differences between men and women were <10%. One study reported that the association between total cholesterol and CHD was higher in men when compared with women [for men HR 2.44, 95% CI 2.00–2.96 per 1 mmol/L increase (n = 20725) vs. women HR 1.93, 95% CI 1.27–2.94 (n = 23525)] without differences in baseline levels.18

Regarding LDL-C, no differences in baseline levels between men and women were reported in any of the eight papers found (see Supplementary material online,Table S3). Two studies reported that the association between LDL-C and composite endpoint/CHD was different in men when compared with women, both pointing in a different direction.19,20 For composite endpoints, women had a higher HR (HR 1.18, 95% CI 1.02–1.37 per SD increase (n = 1625)) when compared with men (HR 1.06, 95% CI 0.94–1.20, n = (1441)) in the Framingham Offspring Study in the USA.20 For the endpoint CHD, a cohort study in Japan reported for women a HR of 1.78 [95% CI 0.66–4.77 for the highest quintile (n = 2525)] and nearly a doubled HR for men, 3.73 (95% CI 1.25–11.1 (n = 2169).19

For HDL-C (13 papers), women were consistently reported to have higher values compared with men15,17,18,21 while the association with all endpoints was similar between men and women (see Supplementary material online,Table S4). Only one study reported a slightly higher HR for CHD in women when compared with men with a lower HDL-C [women HR 1.93, 95% CI 1.27–2.94 for HDL-C <1.03 mmol/L (n = 23525), men HR 1.85, 95% CI 1.53–2.24 (n = 20725)].18

With regards to triglycerides (nine papers), the majority of the literature reported no significant sex differences in baseline triglyceride levels and in their association with CVD endpoints (see Supplementary material online,Table S5). However, one paper reported higher levels in men than women, but despite this difference, the association with CHD was stronger in women compared with men [women HR 1.40, 95% CI 1.15–1.70 per 1 mmol/L increase (n = 9681) vs. men HR 1.21, 95% CI 1.08–1.37 (n = 8888)].16 Another study found that the positive association of triglycerides for composite events was stronger in women than men,22 although this was not statistically significant (women HR 1.73, 95% CI 1.24–2.40 per 1 mmol/L increase, men HR 1.48 95% CI 1.11–1.97 (n = 6395 in total)). When stratified by fasting status, the associations between triglyceride levels and composite events were positive but did not differ between fasting men and non-fasting men. However, the association was more marked for non-fasting women than fasting women [fasting women HR 1.36, 95% CI 0.70–2.64, P = 0.34, non-fasting women HR 1.87 95% CI 1.28–2.73, P < 0.001, the corresponding values for men HR 1.75, 95% CI 1.03–3.07 P = 0.06 and HR 1.34, 95% CI 0.95–1.88, P = 0.02 (n = 4265 in total)].22

One paper reported on total cholesterol to HDL-C ratio, which found that the ratio was not associated with the risk of stroke, nor was there a statistically significant interaction between sex and ischaemic stroke [women PAR 0.13, 95% CI 0.03–0.45 (n = 4112), men HR < 1, PAR NA (n = 2732)].23

For Lp(A) (six papers) half of the papers showed that Lp(A) levels were higher in women when compared with men (see Supplementary material online,Table S6).24–26 For CHD and stroke end points, the evidence was inconsistent as two studies showed that the association between Lp(A) and stroke was stronger in men when compared with women,24,27 yet Ohira et al. showed the opposite for stroke in a white population28 and Kinley et al. showed that the association with CHD was stronger in women when compared with men.26

For ApoA1 (four papers), one study reported higher baseline levels in women when compared with men with no differences in the association with heart failure (see Supplementary material online,Table S7).29

One study reported higher baseline levels of ApoB in men compared with women with no differences in the association with CHD.30

Therefore, apart from LDL-C and HDL-C, differences in reported baseline lipid levels between men and women are inconsistent. When a sex-specific association with the biomarker and outcome was reported in more than one paper, excluding the case of triglycerides where both papers found similar findings, results were conflicting as to which sex had a stronger association.

Biomarkers of inflammation

With regards to c-reactive protein (CRP) (eight papers), three studies reported sex differences in baseline levels with men having higher levels than women (see Supplementary material online,Table S8).31–33 One of these studies reported that the association between CRP and both CHD and composite endpoints was stronger in men when compared with women [for CHD women HR 0.33, 95% CI 0.06–1.88 for highest quartile (n = 266) and men HR 2.39, 95% CI 1.08–5.28 (n = 426)].33 This study also reported that the association between CRP and AF was stronger in men compared with women [men HR 1.14, 95% CI 1.02–1.28 (n = 3093), women HR 0.98, 95% CI 0.85–1.13 (n = 3222)].33

Biomarkers of kidney function

For GFR (six papers), creatinine (two papers), and cystatin C (one paper), there were no reports on sex differences in baseline levels of these kidney markers (see Supplementary material online,Table S9). One study reported that the association of creatinine with composite events was stronger in women than men [women, B = −56.3, p-value = 0.01 (n = 3517) and men B = −20.7, p-value = 0.44 (n = 3126)].34

Glucose and insulin

Sex-specific data on glucose metabolism and its association with CVD endpoints were reported in five publications (see Supplementary material online,Table S10). Biomarkers for glucose metabolism included fasting glucose (three papers), insulin resistance (one paper), and HbA1c (one paper). No publications showed differences at baseline between men and women. The only study on HbA1c and composite outcome showed that the OR was higher in women when compared with men [women OR 2.6, 95% CI 1.1–6.9 (n = 1498) and men OR 1.4, 95% CI 0.7–2.9 (n = 1232)].35 For the association between HbA1c and composite outcome for insulin resistance, homeostasis model assessment (HOMA) was present in a quarter of the general population, and resulted in a more unfavourable risk for stroke in men when compared with women [women RR 1.27, 95% CI 0.41–3.92 for HOMA (Quantile 4) (n = 968) and men RR 10.9, 95% CI 3.04–38.82 (n = 541)].36

Biomarkers of myocyte necrosis

There were two publications on cardiac troponin T, both of which showed that the prevalence of detectable cardiac troponin T was higher in men when compared with women. In the association between the biomarker and the outcome of composite endpoints, De Lemos et al. reported high HRs for both men and women.37 The highest quintile compared with the lowest quintile resulted in a HR of 25.3, 95% CI 10.2–62.8 for men (n = 1565) and an HR of 9.3, 95% CI 2.0–42.1 for women (n = 1981). The authors did not report that this was significantly different between men and women. Saunders et al. found stronger associations between troponin T and composite events in women than men but this was not significantly different. With heart failure as the endpoint, they found a stronger association in women than men but again this was not significantly different. Yet the added clinical value in terms of net reclassification index was higher for women [19.2%, 95% CI 6.7–25.2 (n = 5706)) when compared with men (7.2%, 95% CI 1.9–19.1 (n = 3992)] for composite endpoints and not different for heart failure [women net reclassification index 16.7%, 95% CI 6.4–22.4 (n = 5706) and men net reclassification index 15.9%, 95% CI 8.0–27.0 (n = 3992)].38

Biomarkers of myocyte stress

Sex-specific data on B-type natriuretic peptide (BNP) and its association with CVD was reported in four publications (see Supplementary material online,Table S11). All four studies showed that women had higher levels of BNP at baseline compared with men.39–42 With a different cut-off value for high risk (women 55 pg/mL and men 37 pg/mL), the association between BNP and composite endpoint was stronger in men when compared with women [women HR 1.68, 95% 1.13–2.50 (n = 8844) vs. men HR 3.15, 95% 2.03–4.88 (n = 4365)].39 With AF as an endpoint using cut-off values according to recently published data (women 45 pg/mL and men 31 pg/mL), results did not differ considerably by sex. However, a stronger association was found for males compared with females [women HR 1.34, 95% CI 1.09–1.65 (n = 1597) vs. men HR 1.49, 95% CI 1.23–1.82 (n = 1470)].38

There were no data reports on the predictive value of galectin-3 for CVD in the general population in a sex-stratified manner.

Other biomarkers

The only publication regarding ST2/Interleukin 1 receptor-like 1 (ST2/IL1Rl-1) found that men had higher levels of ST2 than women (men 30.4 ng/mL, women 23.8 ng/mL). Associations between ST2 and composite events, CHD, stroke, and heart failure were comparable between men and women with no interaction between sex and ST2.43

There were two publications regarding vitamin D as a biomarker (see Supplementary material online,Table S12).44,45 Both studies showed that levels of vitamin D were similar between men and women. For CHD, the HR was lower in women when compared with men, yet this was not reported to be different. For stroke as an endpoint, no sex-specific data on baseline levels were presented, and HRs were significant in women, but not in men [women per percentile decrease: HR 1.67, 95% CI 1.30–2.13 (n = 4678) vs. men: HR 1.30, 95% CI 0.97–1.75 (n = 5492)].44 This suggests that the risk of CHD or stroke is higher for women than men with low levels of vitamin D.

For miRNAs, only one study published data on miRNA 126, miRNA 197, miRNA 223, and CHD. Baseline differences between men and women were not published, and the associations with CHD for these miRNAs were of similar magnitude between men and women.46

There were no publications that reported on sex-specific data for GDF-15, or metabolites as biomarkers for CVD in the general population.

The results of the risk of bias assessment for all 54 studies can be found summarized in Supplementary material online, Tables S13 and S14. According to the risk assessment tool, all of the studies had a low or intermediate risk of bias. The poorest performing section for the majority of the studies was the adequacy of follow-up of cohorts in which 28 cohort and case–cohort studies and four case–control studies scored no stars. This was due to there being no indication as to whether or not all subjects had been accounted for or in the cases of subjects being lost to follow-up, the numbers lost were not provided or there was no description of the lost to follow-up group. Therefore, this points to potential bias resulting from loss to follow-up in these studies. Another source of potential bias was regarding representation of the exposed cohorts. Three studies selected a cohort of patients of a specific age range, for example 35–52 years old, which only represents a pre-menopausal population in women.

To fulfil our objectives of investigating the reporting of sex-stratification, we specifically selected publications which had reported sex differences therefore purposefully resulting in selection bias.

Summary

To summarize the data, Table 2 shows that sex differences are reported overall in <50% of the studies. On the whole, there were inconsistencies in the reporting of sex difference in baseline biomarker levels between the studies. The most prominent differences between men and women in baseline biomarker levels relate to the cardiac biomarkers involved in myocyte necrosis and myocyte stress in which men more often have a detectable level of troponin, and women have higher levels of BNP. For the association with cardiovascular endpoints, sex differences were reported in a minority of the studies. Despite sex differences in baseline levels of cardiac biomarkers, there are no significant sex differences in the predictive value of troponin levels for cardiovascular events. For BNP, the data are inconsistent with only one study reporting significantly different associations between the biomarker level and cardiovascular endpoints between men and women with the predictive value being stronger in men. Some papers also show evidence of the biomarkers associating differently in related cardiovascular endpoints. For example, Lp(A) appeared to have a stronger relation with stroke in two studies in men and with stroke and CHD in two studies in women. Two other general population-based studies did not show any differences. In the case of emerging biomarkers, miRNA, GDF-15, and metabolites, differences in baseline levels between men and women were either not published or for the latter two biomarkers, there were no publications reporting on sex-specific data at all. Lack of power of both men and women could account for the inconsistencies in associations found and also in the lack of associations found.

Discussion

Our systematic review on sex-specific differences in established and emerging biomarkers for CVD prediction in primary prevention revealed that a limited number of publications explicitly provide sex-specific information regarding both men and women on the baseline levels of the biomarkers, and on the association of the biomarkers with incident CVD events in the general population. The majority of studies corrected for sex in multivariable models, but do not display sex-specific results. Most of the biomarkers that we retrieved from the literature revealed similar baseline levels between men and women, except for HDL-C, the cardiac biomarkers troponin and BNP, CRP, and ST2. When associations between the biomarker and CVD arose, it was in the same direction for both men and women. Data on biomarkers for AF as an endpoint stratified for sex were also lacking with only one study including it as an endpoint.

CRP is a marker of inflammation involved in the process of atherosclerosis. Higher levels are known to be associated with increased risk of cardiovascular events in asymptomatic individuals.47 In contrast to previous studies in healthy people,48 we found that men have higher CRP levels than women. Accordingly, these men have a higher risk of cardiovascular events than women.33

The most prominent reported differences in baseline biomarker levels related to the cardiac biomarkers. BNP is released from the myocardium in response to myocardial stretch, and is useful in the diagnosis of heart failure when patients present with dyspnoea of unknown origin and to assess response to treatment in patients with diagnosed heart failure. Sex and age-based reference ranges are established from clinical trials and from populations screened for the absence of CVD. In support of what has previously been published in healthy subjects,49 we did find published evidence that BNP levels are consistently higher in women when compared with men in the general population. Yet the predictive value of BNP for CVD appeared to be stronger in men than women. This was with lower cut-off values used for men in two of the papers.39,42

We found that some papers regarding certain biomarkers were in conflict when it came to reporting which sex had the stronger association with the cardiovascular endpoints. Lp(a) plays an important role in atherothrombogenesis, being involved in the initiation and progression of atherosclerosis.50 It has also been found to be associated with endothelial dysfunction51 and may also be involved in the induction of inflammation.52 We found inconsistent results in terms of reporting of associations of lp(a) between men and women with stroke as an endpoint with two studies reporting a significantly stronger predictive value in men and one study reporting a significantly stronger association in women compared with men with the same outcome. Ariyo et al. which found a stronger association in men, and Ohira et al. which found a stronger association in women both used similar sample sizes. The study of Ohira et al. only looked at a white population and 97.5% of the population included in the study of Ariyo et al. were white. The average population age of the study by Ohira et al. was significantly younger (54 years vs. 72 and 73 years for women and men, respectively). The second study, which found a significantly stronger association in men compared with women, also used a cohort of a similar age to the study of Ariyo et al. Therefore, one plausible reason behind the inconsistent results found could be due to differences in ages, suggesting that the atherothrombotic effects of lp(a) are age specific, in keeping with previous literature.53

The year of publication of the papers included in this review date back to as early as 1990. One possible explanation for the inconsistent results found in the associations between men and women and the conflicting results of CRP with previously known literature could be due to the timing of the paper. For example, the study of Tracey et al reporting higher CRP in men than women was published back in 1997. CVD has changed dramatically over the years including the use of different diagnostic assays being used therefore it is important to consider.

Some of the studies showed the potential for follow-up and lack of response bias occurring. This can lead to overestimation or underestimation of any associations found so should be taken into account as can be a cause for the inconsistent or lack of associations found between the biomarkers and outcome in men and women.

Insufficiently powered studies, of both men and women could account for the inconsistencies in associations found between the biomarkers and CVD between the sexes and should also be taken into account in the cases where a lack of association was found.

We report that there were no sex-specific data available on emerging biomarkers such as GDF-15, metabolomics, and miRNAs. This may be explained by the fact that many new biomarkers are costly, and their measurements are often not yet standardized or simply not feasible to measure in many individuals. However, early animal models and smaller studies in humans indicate that metabolomics and miRNAs do display significant differences in sex. When considering these emerging biomarkers for risk prediction, sex differences need to be examined.

Rigorous statistical testing that provides information on the added clinical value of novel biomarkers on top of existing risk factors includes reclassification analyses. Such analyses performed separately in men and women on top of current risk prediction models were only reported in a sex-specific manner in seven studies. Established biomarkers measured in large cohorts including troponins and BNP all show evidence of sex-specific differences. Given the need for biomarkers to be cost-effective and to meet the demands associated with rising healthcare costs, it is essential therefore, to establish clinical utility separately for men and women and to compare these to other related biomarkers, in populations with different levels of absolute risk.

Cardiovascular disease research has predominantly been performed in men. One explanation which could account for this discrepancy is the notion that women tend to develop CVD ∼10 years later than men and subsequently are not eligible for many trials on, for example heart failure. Our systematic review indicates that women are not underrepresented in general population-based cohorts, as the majority of papers included as many women as men. However, many studies used sex in multivariable analyses but did not stratify their analyses by sex. This may be due to non-significant interaction terms for sex, which did not justify subgroup analyses from a statistical point of view. However, many high ranked journals such as The Lancet still encourage analysing data by sex and race (guide for authors, The Lancet).

The majority of the studies included men and women of a similar age. However, as women are affected by CVD at an older age than men it may also result in differences in the associations seen between the biomarkers and the outcomes which may explain the inconsistencies in the reported associations. Public attitudes regarding CVD as mainly a problem for men have been challenged successfully through media campaigns in the general population.54 Recognition of important sex differences in CVD prevention has led to the formulation of specific guidelines for women. However, the American Heart Association guidelines for CVD prevention in women differ little from the guidelines for men. Our data support the lack of differences in established biomarkers comprising lipoproteins, cholesterol parameters among others between men and women for CVD prediction. Yet, Framingham risk score still performs poorly in women when compared with men, especially of older age.55 This underscores the need for investigating sex-specific differences in novel biomarkers.

Limitations

We were unable to perform a meta-analysis of results as one unit increase for biomarkers used to assess the association with CVD varied by study. Some studies reported percentiles, whereas other studies reported categories of which cut-off values differed.

As we specifically selected articles that provided sex-stratified data, some selection bias seems likely. Many population-based cohorts have included only men or only women. An example is the landmark Nurses' Health Study that has reported on many biomarkers for CVD in women.56 Yet, this and other large cohort studies with either men or women were not included in this review as we set out to directly compare the baseline levels of biomarkers and their predictive capacity between sexes that were detected on identical platforms.

Insufficiently powered studies may lead to the inability of the authors to be able to detect differences in associations. This is particularly relevant when stratifying studies by sex as it invariably leads to a reduction of numbers in each group investigated.

In summary, sex-specific data on biomarkers for CVD in the general population exist in large cohorts, but is underreported. Instead of using sex as risk factor in multivariable models, sex interactions should be performed prior to the stratification by sex if the total numbers of the individual groups allow and therefore sex-specific data reported if stratification is justified. Of the evidence available, there is inconsistency in sex-specific differences in levels of traditional biomarkers and their relation to CVD. In order to improve cardiovascular care in men and women, reporting sex-specific data on clinical utility of biomarkers is crucial and should be a requirement in publications of sufficiently powered studies.

Supplementary material

Supplementary material is available at European Heart Journal – Quality of Care and Clinical Outcomes online.

Funding

1) A.G. is supported by EUTRAIN (European Translational tRaining for Autoimmunity & Immune manipulation Network). This project has received funding from the 7th Framework programme of the EU, SP3-People, support for training and career development for researchers (Marie Curie), Network for Initial Training (ITN), FP7-PEOPLE-2011-ITN, under the Marie Sklodowska-Curie grant agreement No. 289903. 2) Dutch Heart Foundation 2013T084 “Queen of Hearts”.

References

1
Gholizadeh
L
,
Davidson
P
.
More similarities than differences: an international comparison of CVD mortality and risk factors in women
.
Health Care Women Int
 
2008
;
29
:
3
22
.
2
Mozaffarian
D
,
Benjamin
EJ
,
Go
AS
,
Arnett
DK
,
Blaha
MJ
,
Cushman
M
,
de Ferranti
S
,
Després
J-P
,
Fullerton
HJ
,
Howard
VJ
,
Huffman
MD
,
Judd
SE
,
Kissela
BM
,
Lackland
DT
,
Lichtman
JH
,
Lisabeth
LD
,
Liu
S
,
Mackey
RH
,
Matchar
DB
,
McGuire
DK
,
Mohler
ER
3rd
,
Moy
CS
,
Muntner
P
,
Mussolino
ME
,
Nasir
K
,
Neumar
RW
,
Nichol
G
,
Palaniappan
L
,
Pandey
DK
,
Reeves
MJ
,
Rodriguez
CJ
,
Sorlie
PD
,
Stein
J
,
Towfighi
A
,
Turan
TN
,
Virani
SS
,
Willey
JZ
,
Woo
D
,
Yeh
RW
,
Turner
MB
;
on behalf of the American Heart Association Statistics Committee and Stroke Statistics Subcommittee
.
Heart disease and stroke statistics—2015 update: a report from the American Heart Association
.
Circulation
 
2015
;
131
:
e29
e322
.
3
Berger
JS
,
Jordan
CO
,
Lloyd-Jones
D
,
Blumenthal
RS
.
Screening for cardiovascular risk in asymptomatic patients
.
J Am Coll Cardiol
 
2010
;
55
:
1169
1177
.
4
Arbustini
E
,
Dal Bello
B
,
Morbini
P
,
Burke
AP
,
Bocciarelli
M
,
Specchia
G
,
Virmani
.
Plaque erosion is a major substrate for coronary thrombosis in acute myocardial infarction
.
Heart
 
1999
;
82
:
269
272
.
5
Hellings
WE
,
Pasterkamp
G
,
Verhoeven
BAN
,
De Kleijn
DP
,
De Vries
JP
,
Seldenrijk
KA
,
van den Broek
T
,
Moll
FL
.
Gender-associated differences in plaque phenotype of patients undergoing carotid endarterectomy
.
J Vasc Surg
 
2007
;
45
:
289
296
;
discussion 296–297
.
6
Vrijenhoek
JEP
,
Den Ruijter
HM
,
De Borst
GJ
,
De Kleijn
DP
,
De Vries
JP
,
Bots
ML
,
Van de Weg
SM
,
Vink
A
,
Moll
FL
,
Pasterkamp
G
.
Sex is associated with the presence of atherosclerotic plaque hemorrhage and modifies the relation between plaque hemorrhage and cardiovascular outcome
.
Stroke
 
2013
;
44
:
3318
3323
.
7
Den Ruijter
H
,
Pasterkamp
G
,
Rutten
FH
,
Lam
CSP
,
Chi
C
,
Tan
KH
,
van Zonneveld
AJ
,
Spaanderman
M
,
de Kleijn
DPV
.
Heart failure with preserved ejection fraction in women: the Dutch Queen of Hearts program
.
Netherlands Heart J
 
2015
;
23
:
89
93
.
8
Gijsberts
CM
,
Gohar
A
,
Ellenbroek
GHJM
,
Hoefer
IE
,
de Kleijn
DPV
,
Asselbergs
FW
,
Nathoe
HM
,
Agostoni
P
,
Rittersma
S
,
Pasterkamp
G
,
Appelman
Y
,
den Ruijter
HM
.
Severity of stable coronary artery disease and its biomarkers differ between men and women undergoing angiography
.
Atherosclerosis
 
2015
;
241
:
234
240
.
9
Peters
SAE
,
Huxley
RR
,
Woodward
M
.
Diabetes as a risk factor for stroke in women compared with men: a systematic review and meta-analysis of 64 cohorts, including 775,385 individuals and 12,539 strokes
.
The Lancet
 
2014
;
383
:
1973
1980
.
10
Ko
DT
,
Wijeysundera
HC
,
Udell
JA
,
Vaccarino
V
,
Austin
PC
,
Guo
H
,
Velianou
JL
,
Lau
K
,
Tu
JV
.
Traditional cardiovascular risk factors and the presence of obstructive coronary artery disease in men and women
.
Can J Cardiol
 
2014
;
30
:
820
826
.
11
Mosca
L
,
Benjamin
EJ
,
Berra
K
,
Bezanson
JL
,
Dolor
RJ
,
Lloyd-Jones
DM
,
Newby
LK
,
Pina
IL
,
Roger
VL
,
Shaw
LJ
,
Zhao
D
,
Beckie
TM
,
Bushnell
C
,
D'Armiento
J
,
Kris-Etherton
PM
,
Fang
J
,
Ganiats
TG
,
Gomes
AS
,
Gracia
CR
,
Haan
CK
,
Jackson
EA
,
Judelson
DR
,
Kelepouris
E
,
Lavie
CJ
,
Moore
A
,
Nussmeier
NA
,
Ofili
E
,
Oparil
S
,
Ouyang
P
,
Pinn
VW
,
Sherif
K
,
Smith
SC
Jr
,
Sopko
G
,
Chandra-Strobos
N
,
Urbina
EM
,
Vaccarino
V
,
Wenger
NK
.
Effectiveness-based guidelines for the prevention of cardiovascular disease in women-2011 update: a Guideline from the American Heart Association
.
Circulation
 
2011
;
123
:
1243
1262
.
12
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
8
.
13
Zeller
T
,
Hughes
M
,
Tuovinen
T
,
Schillert
A
,
Conrads-Frank
A
,
Ruijter
HD
,
Schnabel
RB
,
Kee
F
,
Salomaa
V
,
Siebert
U
,
Thorand
B
,
Ziegler
A
,
Breek
H
,
Pasterkamp
G
,
Kuulasmaa
K
,
Koenig
W
,
Blankenberg
S
.
BiomarCaRE: rationale and design of the European BiomarCaRE project including 300,000 participants from 13 European countries
.
Eur J Epidemiol
 
2014
;
29
:
777
790
.
14
Institute
OHR
.
The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses
. .
15
Houterman
S
,
Verschuren
WM
,
Hofman
A
,
Witteman
JC
.
Serum cholesterol is a risk factor for myocardial infarction in elderly men and women: the Rotterdam Study
.
J Intern Med
 
1999
;
246
:
25
33
.
16
Jónsdóttir
LS
,
Sigfússon
N
,
Gudnason
V
,
Sigvaldason
H
,
Thorgeirsson
G
.
Do lipids, blood pressure, diabetes, and smoking confer equal risk of myocardial infarction in women as in men? The Reykjavik Study
.
J Cardiovasc Risk
 
2002
;
9
:
67
76
.
17
Njolstad
I
,
Arnesen
E
,
Lund-Larsen
PG
.
Smoking, serum lipids, blood pressure, and sex differences in myocardial infarction: a 12-year follow-up of the Finnmark Study
.
Circulation
 
1996
;
93
:
450
456
.
18
Madssen
E
,
Laugsand
LE
,
Wiseth
R
,
Mørkedal
B
,
Platou
C
,
Vatten
L
,
Janszky
I
.
Risk of acute myocardial infarction: dyslipidemia more detrimental for men than women
.
Epidemiology
 
2013
;
24
:
637
642
.
19
Okamura
T
,
Kokubo
Y
,
Watanabe
M
,
Higashiyama
A
,
Ono
Y
,
Miyamoto
Y
,
Yoshimasa
Y
,
Okayama
A
.
Triglycerides and non-high-density lipoprotein cholesterol and the incidence of cardiovascular disease in an urban Japanese cohort: the Suita study
.
Atherosclerosis
 
2009
;
209
:
290
294
.
20
Cromwell
WC
,
Otvos
JD
,
Keyes
MJ
,
Michael
J
,
Sullivan
L
,
Vasan
RS
,
Wilson
PWF
,
D'Agostino
RB
.
LDL particle number and risk of future cardiovascular disease in the Framingham Offspring Study – Implications for LDL management
.
J Clin Lipidol
 
2007
;
1
:
583
592
.
21
Cooney
MT
,
Dudina
A
,
De Bacquer
D
,
Wilhelmsen
L
,
Sans
S
,
Menotti
A
,
De Backer
G
,
Jousilahti
P
,
Keil
U
,
Thomsen
T
,
Whincup
P
,
Graham
JM
.
HDL cholesterol protects against cardiovascular disease in both genders, at all ages and at all levels of risk
.
Atherosclerosis
 
2009
;
206
:
611
616
.
22
Iso
H
,
Imano
H
,
Yamagishi
K
,
Ohira
T
,
Cui
R
,
Noda
H
,
Sato
S
,
Kiyama
M
,
Okada
T
,
Hitsumoto
S
,
Tanigawa
T
,
Kitamura
A
.
Fasting and non-fasting triglycerides and risk of ischemic cardiovascular disease in Japanese men and women: the Circulatory Risk in Communities Study (CIRCS)
.
Atherosclerosis
 
2014
;
237
:
361
368
.
23
Bos
MJ
,
Koudstaal
PJ
,
Hofman
A
,
Ikram
MA
.
Modifiable etiological factors and the burden of stroke from the Rotterdam study: a population-based cohort study
.
PLoS Med
 
2014
;
11
:
e1001634
.
24
Ariyo
AA
,
Thach
C
,
Tracy
R
.
Lp(a) lipoprotein, vascular disease, and mortality in the elderly
.
N Engl J Med
 
2003
;
349
:
2108
2115
.
25
Ishikawa
S
,
Kotani
K
,
Kario
K
,
Kayaba
K
,
Gotoh
T
,
Nakamura
Y
,
Kajii
E
.
Inverse association between serum lipoprotein(a) and cerebral hemorrhage in the Japanese population
.
Thromb Res
 
2013
;
131
:
e54
e58
.
26
Kinlay
S
,
Dobson
AJ
,
Heller
RF
,
McElduff
P
,
Alexander
H
,
Dickeson
J
.
Risk of primary and recurrent acute myocardial infarction from lipoprotein(a) in men and women
.
JACC
 
1996
;
28
:
1390
1397
.
27
Boden-Albala
B
,
Kargman
DE
,
Lin
IF
,
Paik
MC
,
Sacco
RL
,
Berglund
L
.
Increased stroke risk and lipoprotein(a) in a multiethnic community: the northern Manhattan stroke study
.
Cerebrovasc Dis
 
2010
;
30
:
237
243
.
28
Ohira
T
,
Schreiner
PJ
,
Morrisett
JD
,
Chambless
LE
,
Rosamond
WD
,
Folsom
AR
.
Lipoprotein(a) and incident ischemic stroke: the Atherosclerosis Risk in Communities (ARIC) study
.
Stroke
 
2006
;
37
:
1407
1412
.
29
Holme
I
,
Aastveit
AH
,
Hammar
N
,
Jungner
I
,
Walldius
G
.
Lipoprotein components and risk of congestive heart failure in 84 740 men and women in the Apolipoprotein MOrtality RISk study (AMORIS)
.
Eur J Heart Fail
 
2009
;
11
:
1036
1042
.
30
Meisinger
C
,
Loewel
H
,
Mraz
W
,
Koenig
W
.
Prognostic value of apolipoprotein B and A-I in the prediction of myocardial infarction in middle-aged men and women: results from the MONICA/KORA Augsburg cohort study
.
Eur Heart J
 
2005
;
26
:
271
278
.
31
Tracy
RP
,
Lemaitre
RN
,
Psaty
BM
,
Ives
DG
,
Evans
RW
,
Cushman
M
,
Meilahn
EN
,
Kuller
LH
.
Relationship of C-reactive protein to risk of cardiovascular disease in the elderly: results from the cardiovascular health study and the rural health promotion project
.
Arterioscler Thromb Vasc Biol
 
1997
;
17
:
1121
1127
.
32
Tuomisto
K
,
Jousilahti
P
,
Sundvall
J
,
Pajunen
P
,
Salomaa
V
,
Bounameaux
H
.
C-reactive protein, interleukin-6 and tumor necrosis factor alpha as predictors of incident coronary and cardiovascular events and total mortality A population-based, prospective study
.
Thromb Haemost
 
2006
;
95
:
56
64
.
33
Nyrnes
A
,
Njolstad
I
,
Mathiesen
EB
,
Wilsgaard
T
,
Hansen
JB
,
Skjelbakken
T
,
Jørgensen
L
,
Løche
LM
.
Inflammatory biomarkers as risk factors for future atrial fibrillation. An eleven-year follow-up of 6315 men and women: The Tromsø study
.
Gend Med
 
2012
;
9
:
536
547.e2
.
34
Ito
H
,
Pacold
IV
,
Durazo-Arvizu
R
,
Liu
K
,
Shilipak
MG
,
Goff
DC
,
Tracy
RP
,
Kramer
H
.
The effect of including cystatin C or creatinine in a cardiovascular risk model for asymptomatic individuals
.
Am J Epidemiol
 
2011
;
174
:
949
957
.
35
Adams
RJ
,
Appleton
SL
,
Hill
CL
,
Wilson
DH
,
Taylor
AW
,
Chittleborough
CR
,
Gill
TK
,
Ruffin
RE
.
Independent association of HbA(1c) and incident cardiovascular disease in people without diabetes
.
Obesity (Silver Spring)
 
2009
;
17
:
559
563
.
36
Rundek
T
,
Gardener
H
,
Xu
Q
,
Goldberg
RB
,
Wright
CB
,
Boden-Albala
B
,
Disla
BS
,
Paik
MC
,
Elkind
MSV
,
Sacco
RL
.
Insulin resistance and risk of ischemic stroke among nondiabetic individuals from the northern Manhattan study
.
Arch Neurol
 
2010
;
67
:
1195
1200
.
37
De Lemos
JA
,
Drazner
MH
,
Morrow
DA
.
Association of troponin T detected with a highly sensitive assay and cardiac structure and mortality risk in the general population
.
JAMA
 
2010
;
304
:
2503
2512
.
38
Saunders
JT
,
Nambi
V
,
De Lemos
JA
,
Chambless
LE
,
Virani
SS
,
Boerwinkle
E
,
Hoogeveen,
RC
,
Liu
X
,
Astor
BC
,
Mosley
TH
,
Folsom
AR
,
Heiss
G
,
Coresh
J
,
Ballantyne
CM
.
Cardiac troponin T measured by a highly sensitive assay predicts coronary heart disease, heart failure, and mortality in the atherosclerosis risk in communities study
.
Circulation
 
2011
;
123
:
1367
1376
.
39
Nakamura
M
,
Tanaka
F
,
Takahashi
T
,
Makita
S
,
Ishisone
T
,
Onodera
M
,
Ishibashi,
Y
,
Itai
K
,
Onoda
T
,
Ohsawa
M
,
Tanno
K
,
Sakata
K
,
Shinichi
O
,
Ogasawara
K
,
Ogawa
A
,
Kuribayashi
T
,
Okayama
A
.
Sex-specific threshold levels of plasma B-type natriuretic peptide for prediction of cardiovascular event risk in a Japanese population initially free of cardiovascular disease
.
Am J Cardiol
 
2011
;
108
:
1564
1569
.
40
Rutten
JHW
,
Mattace-Raso
FUS
,
Steyerberg
EW
,
Lindemans
J
,
Hofman
A
,
Wieberdink
RG
,
Breteler
MB
,
Witteman
JCM
,
van den Meiracker
AH
.
Amino-terminal Pro-B-type natriuretic peptide improves cardiovascular and cerebrovascular risk prediction in the population: The rotterdam study
.
Hypertension
 
2010
;
55
:
785
791
.
41
Takahashi
T
,
Nakamura
M
,
Onoda
T
,
Ohsawa
M
,
Tanno
K
,
Itai
K
,
Sakata
K
,
Sakuma
M
,
Tanaka
F
,
Makita
S
,
Yoshida
Y
,
Ogawa
A
,
Kawamura
K
,
Okayama
A
.
Predictive value of plasma B-type natriuretic peptide for ischemic stroke: a community-based longitudinal study
.
Atherosclerosis
 
2009
;
207
:
298
303
.
42
Kara
K
,
Geisel
MH
,
Möhlenkamp
S
,
Lehmann
N
,
Kälsch
H
,
Bauer
M
,
Neumann
T
,
Dragano
N
,
Moebus
S
,
Jockel
KH
,
Erbel
R
,
Mahabadi
AA
.
B-type natriuretic peptide for incident atrial fibrillation – the Heinz Nixdorf Recall Study
.
J Cardiol
 
2014
;
65
:
453
458
.
43
Hughes
MF
,
Appelbaum
S
,
Havulinna
AS
,
Jagodzinski
A
,
Zeller
T
,
Kee
F
,
Blankenberg
S
,
Salomaa
V
.
ST2 may not be a useful predictor for incident cardiovascular events, heart failure and mortality
.
Heart
 
2014
;
100
:
1715
1721
.
44
Brøndum-Jacobsen
P
,
Nordestgaard
BG
,
Schnohr
P
,
Benn
M
.
25-Hydroxyvitamin D and symptomatic ischemic stroke: an original study and meta-analysis
.
Ann Neurol
 
2013
;
73
:
38
47
.
45
Karakas
M
,
Thorand
B
,
Zierer
A
,
Huth
C
,
Meisinger
C
,
Roden
M
,
Rottbauer
W
,
Peters
A
,
Koenig
W
,
Herder
C
.
Low levels of serum 25-hydroxyvitamin D are associated with increased risk of myocardial infarction, especially in women: results from the MONICA/KORA Augsburg case-cohort study
.
J Clin Endocrinol Metab
 
2012
;
97
:
4578
4587
.
46
Zampetaki
A
,
Willeit
P
,
Tilling
L
,
Drozdov
I
,
Prokopi
M
,
Renard
JM
,
Mayr
A
,
Weger
S
,
Schett
G
,
Shah
A
,
Boulanger
CM
,
Willeit
J
,
Chowienczyk
PJ
,
Kiechl
S
,
Mayr
M
.
Prospective study on circulating microRNAs and risk of myocardial infarction
.
J Am Coll Cardiol
 
2012
;
60
:
290
299
.
47
Vivona
N
,
Bivona
G
,
Noto
D
,
Sasso
BL
,
Cefalù
AB
,
Chiarello
G
,
Falletta
A
,
Ciaccio
M
,
Averna
MR
.
C-reactive protein but not soluble CD40 ligand and homocysteine is associated to common atherosclerotic risk factors in a cohort of coronary artery disease patients
.
Clin Biochem
 
2009
;
42
:
1713
1718
.
48
Wong
ND
,
Pio
J
,
Valencia
R
,
Thakal
G
.
Distribution of C-reactive protein and its relation to risk factors and coronary heart disease risk estimation in the National Health and Nutrition Examination Survey (NHANES) III
.
Prev Cardiol
 
2001
;
4
:
109
114
.
49
Clerico
A
,
Del Ry
S
,
Maffei
S
,
Prontera
C
,
Emdin
M
,
Giannessi
D
.
The circulating levels of cardiac natriuretic hormones in healthy adults: effects of age and sex
.
Clin Chem Lab Med
 
2002
;
40
:
371
377
.
50
Loscalzo
J
.
Review lipoprotein(a): a unique risk factor for atherothrombotic disease
.
Arterioscler Thromb Vasc Biol
 
1963
;
10
:
672
679
.
51
Schlaich
MP
,
John
S
,
Langenfeld
MRW
,
Lackner
KJ
,
Schmitz
G
,
Schmieder
RE
.
Does lipoprotein (a) impair endothelial function?
JACC
 
1998
;
31
:
359
365
.
52
Dangas
G
,
Mehran
R
,
Harpel
PC
,
Sharma
SK
,
Marcovina
SM
,
Dube
G
,
Ambrose
JA
,
Fallon
JT
.
Lipoprotein(a) and inflammation in human coronary atheroma: association with the severity of clinical presentation
.
J Am Coll Cardiol
 
1998
;
32
:
2035
2042
.
53
Sunayama
S
,
Daida
H
,
Mokuno
H
,
Miyano
H
,
Yokoi
H
,
Lee
YJ
,
Sakurai
H
,
Yamaguchi
H
.
Lack of increased coronary atherosclerotic risk due to elevated lipoprotein(a) in women > or = 55 years of age
.
Circulation
 
1996
;
94
:
1263
1268
.
54
Mosca
L
,
Barrett-Connor
E
,
Kass Wenger
N
.
Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes
.
Circulation
 
2011
;
124
:
2145
2154
.
55
Rodondi
N
,
Locatelli
I
,
Aujesky
D
,
Butler
J
,
Vittinghoff
E
,
Simonsick
E
,
Satterfield
S
,
Newman
AB
,
Wilson
PWF
,
Pletcher
MJ
,
Bauer
DC
;
for the Health ABC Study
.
Framingham risk score and alternatives for prediction of coronary heart disease in older adults
.
PLoS ONE
 
2012
;
7
:
e34287
.
56
Sun
Q
,
Kiernan
UA
,
Shi
L
,
Phillips
DA
,
Kahn
BB
,
Hu
FB
,
Manson
JE
,
Albert
CM
,
Rexrode
KM
.
Plasma retinol-binding protein 4 (RBP4) levels and risk of coronary heart disease: a prospective analysis among women in the nurses’ health study
.
Circulation
 
2013
;
127
:
1938
1947
.