Vascular age to determine cardiovascular disease risk: A systematic review of its concepts,definitions,and clinical applications

Abstract

Background

Vascular age is an alternate means of representing an individual's cardiovascular risk. Little consensus exists on what vascular age represents and its clinical utility has not been determined. We systematically reviewed the literature to provide a comprehensive overview of different methods that have been used to define vascular age, and to examine its potential clinical value in patient communication and risk prediction.

Design

This was a systematic review with data sources of PubMed and Embase.

Results

We identified 39 articles on vascular age, 20 proposed to use vascular age as a communication tool and 19 proposed to use vascular age as a means to improve cardiovascular risk prediction. Eight papers were methodological and 31 papers reported on vascular age in study populations. Of these 31 papers, vascular age was a direct translation of the absolute risk estimated by existing cardiovascular risk prediction models in 15 papers, 12 derived vascular age from the reference values of an additional test, and in three papers vascular age was defined as the age at which the estimated cardiovascular risk equals the risk from non-invasive imaging observed degree of atherosclerosis. One trial found a small effect on risk factor levels when vascular age was communicated instead of cardiovascular risk.

Conclusion

Despite sharing a common name, various studies have proposed distinct ways to define and measure vascular age. Studies into the effects of vascular age as a tool to improve cardiovascular risk prediction or patient communication are scarce but will be required before its clinical use can be justified.

Keywords

Cardiovascular disease vascular age risk communication risk prediction atherosclerosis

Introduction

Cardiovascular diseases (CVDs) are a leading cause of morbidity and mortality worldwide.¹ The risk of developing symptomatic CVD varies greatly between individuals and can be predicted using various prediction models. Based on the predicted risk for CVD in the next 10 years, an individual can be classified into low, intermediate or high risk categories with their corresponding treatment options.² However, CVD risk is primarily driven by age and a young person with a significant elevated risk factor burden is still likely to have a low CVD risk score. Continuing exposure to high risk factor levels in these young individuals will likely lead to increased absolute CVD risk as they age. Therefore, early identification of high risk individuals when they are young and the implementation of risk factor modification, especially through lifestyle change, would have important health benefits.

Vascular age is a means of expressing cardiovascular risk as an age. It is thought to improve cardiovascular risk prediction models and may contribute to a better understanding of cardiovascular risk, especially as in young patients the long term effects of high risk factor levels can be concealed by a young age. Previous research indicates that vascular age is easily understood by patients and has a greater impact on care than presenting an estimated CVD risk score.³ Furthermore, results from a randomized controlled trial have indicated that targeted communication about 10-year vascular risk leads to improved control of lipid levels.⁴ Recent Canadian guidelines for the diagnosis and treatment of dyslipidaemia have recommended calculating and discussing a patient’s vascular age in order to improve the likelihood that patients will reach targeted lipid levels and better control their hypertension. This has shown to be especially useful, for young individuals where calculated 10-year risk scores are low.⁵

Despite growing interest in vascular age, there is little consensus about the definitions of vascular age and alternative terms such as ‘heart age’ or ‘risk age’ have also been used. We systematically reviewed the current literature on vascular age to provide a comprehensive overview of the approaches that have been used to define and calculate vascular age, and to examine its potential value for risk prediction and improved patient communication.

Methods

Search strategy

PubMed and Embase databases were searched on 27 October 2014, using a search syntax that consisted of combinations of synonyms and plurals of ‘vascular age’ (‘heart age’, ‘risk age’, ‘coronary age’) and was approved by a medical information specialist (Table 1).

Table 1.

Search syntax.

Database	Syntax
PubMed	'vascular age' [TIAB] OR ‘vascular ages' [TIAB] OR ‘vascular ageing' [TIAB] OR ‘vascular aging' [TIAB] OR ‘vessel age' [TIAB] OR ‘vessel aging' [TIAB] OR ‘arterial age' [TIAB] OR ‘arterial ageing' [TIAB] OR ‘arterial aging' [TIAB] OR ‘risk age' [TIAB] OR ‘risk ages' [TIAB] OR ‘heart age' [TIAB] OR ‘heart ages' [TIAB] OR ‘heart aging' [TIAB] OR ‘coronary age' [TIAB] OR ‘cardiovascular age' [TIAB] OR ‘cardiovascular ages' [TIAB] OR ‘cardiovascular ageing' [TIAB] OR ‘cardiovascular aging' [TIAB]
Embase	'vascular age': ab, ti OR ‘vascular ages': ab, ti OR ‘vascular ageing': ab, ti OR ‘vascular aging': ab, ti OR ‘vessel age': ab, ti OR ‘vessel aging': ab, ti OR ‘arterial age': ab, ti OR ‘arterial ageing': ab, ti OR ‘arterial aging': ab, ti OR ‘risk age': ab, ti OR ‘risk ages': ab, ti OR ‘heart age': ab, ti OR ‘heart ages': ab, ti OR ‘heart aging': ab, ti OR ‘coronary age': ab, ti OR ‘cardiovascular age': ab, ti OR ‘cardiovascular ages': ab, ti OR ‘cardiovascular ageing': ab, ti OR ‘cardiovascular aging': ab, ti

Database

Syntax

PubMed

'vascular age' [TIAB] OR ‘vascular ages' [TIAB] OR ‘vascular ageing' [TIAB] OR ‘vascular aging' [TIAB] OR ‘vessel age' [TIAB] OR ‘vessel aging' [TIAB] OR ‘arterial age' [TIAB] OR ‘arterial ageing' [TIAB] OR ‘arterial aging' [TIAB] OR ‘risk age' [TIAB] OR ‘risk ages' [TIAB] OR ‘heart age' [TIAB] OR ‘heart ages' [TIAB] OR ‘heart aging' [TIAB] OR ‘coronary age' [TIAB] OR ‘cardiovascular age' [TIAB] OR ‘cardiovascular ages' [TIAB] OR ‘cardiovascular ageing' [TIAB] OR ‘cardiovascular aging' [TIAB]

Embase

'vascular age': ab, ti OR ‘vascular ages': ab, ti OR ‘vascular ageing': ab, ti OR ‘vascular aging': ab, ti OR ‘vessel age': ab, ti OR ‘vessel aging': ab, ti OR ‘arterial age': ab, ti OR ‘arterial ageing': ab, ti OR ‘arterial aging': ab, ti OR ‘risk age': ab, ti OR ‘risk ages': ab, ti OR ‘heart age': ab, ti OR ‘heart ages': ab, ti OR ‘heart aging': ab, ti OR ‘coronary age': ab, ti OR ‘cardiovascular age': ab, ti OR ‘cardiovascular ages': ab, ti OR ‘cardiovascular ageing': ab, ti OR ‘cardiovascular aging': ab, ti

[TIAB] and ab, ti in Table 1 were codes used to restrict the search to title and abstract fields.

Screening and selection

After excluding exact duplicates, articles were screened on title and abstract. Studies were excluded if they reported on non-human experiments (in vitro, animal studies), were performed in children, did not determine vascular age or the method of defining vascular age was unclear. Secondary sources (reviews, editorials), conference abstracts and papers published in another language than English were also excluded. After title and abstract screening, all retrievable full text articles were screened using the same criteria as were used for the screening of the abstracts, with the addition that the method of defining or measuring vascular age (or synonym) was described. When multiple papers reported on the same cohort, we selected the paper with the largest number of participants. When many groups were mentioned, we calculated averages for more broad categories.

Data extraction

We extracted, whenever appropriate, data on study characteristics, definition of vascular age, proposed methods of defining vascular age, chronological and vascular age in the study, and the yield in cardiovascular risk prediction.

Results

Search strategy and study selection

Our systematic search resulted in 1965 unique articles of which 107 articles were eligible for full text screening. Of these 107 articles, 70 were excluded. From these 70 articles, six were not available in full text, 38 were conference abstracts, nine were not in English, five were secondary sources, six did not determine vascular age, one did not describe methodology and five were multiple reports on the same cohort.

Screening of the references yielded another two articles. Of the 39 selected articles, eight articles reported on concepts and methodologies of vascular age and 31 applied vascular age in a clinical setting. The flowchart of the search and selection process is provided in Figure 1. Characteristics of the included studies where vascular age was applied to a population are presented in Tables 2 and 3. Table 4 contains the results of the studies that examined the association between vascular age and cardiovascular disease.

Figure 1.

Flowchart of study selection.

Table 2.

Details of studies that used vascular age in the context risk communication.

First author	Time of data-collection	Country	Study size (% women)	Population	Mean chronological age	Mean vascular age
Bonner⁶	2012	Australia	26	With cardiovascular risk factors, no medication	54	61
Daher⁷	2008	USA	53 (74)	Cancer survivors	46	54
Davies⁸	2008–2011	UK	749 (33)	HIV-positive adults	42	48 or 60^a
De Socio⁹	2007	Italy	72 (21)	HIV-patients	42	49
García-Ortiz¹⁰	NR	Spain	1365 (60)	General population; no CVD	55	57
Hollar¹¹	2001^b	USA	2654	General population (NHANES), reported by disabilities	NR (differentials reported)	No disability: –14.5 Any disability: –12
Iancu¹²	2011	Romania	47 (55)	Patients with obesity, without CVD, undergoing bariatric surgery	40	Preoperatively: 66 12 months postop: 41
Kakinami¹³	2004–2009	USA	Cases: 588 (35) Controls:	Patients with HIV and or HCV/controls sampled from NHANES	Cases: 48 Controls: 48	Cases: 57 Controls: 54
Lopez-Gonzales¹⁴	2009–2011	Spain	3153 (52)	Public sector workers, participating in trial on effect of way of communicating risk on risk factors	Control: 45 FRS: 46 HA: 47	Control: 49 (1.2) FRS: 50 (–0.3) HA: 50 (–1.5)^c
Neufingerl¹⁵	2009–2011	Multiple	2,744,091 (56)	Users of online ‘heart age’ tool	43	47
Ratliff¹⁶	2008	USA	106/197 (61/57)	Schizophrenic outpatients/matched controls.	Cases: 48 Controls: 48	Cases: 62 Controls: 55
Sharma¹⁷	NR	India	2483 (40)	Apparently healthy individuals	47	53
Soares¹⁸	NR	Brazil	70 (71)	Participants with at least three risk factors, participating in trial of effect of lifestyle changes on FRS	51	Before: 68 After: 65
Tay¹⁹	2010–2011	Singapore	326 (43) 83 cases and 243 controls	Adults with schizophrenia and community-dwelling controls	Cases: 36 Controls: 35	Cases: 41 Controls: 36
Zhang²⁰	NR	France	4697 (NR)^d	Patients with ‘not-at-goal’-hypertension	59	81

CVD: cardiovascular disease; FRS: Framingham risk score; HA: heart age; HCV: hepatitis C virus; HIV: human immunodeficiency virus; NHANES: National Health and Nutrition Examination Survey; NR: not reported.

Smoking status was not available; heart age of the cohort was estimated twice. The first measure is calculated assuming all participants were non-smokers, the second measure assumed all participants were smokers.

This paper reported on multiple cohorts of NHANES. For readability purposes, we present only data from one cohort, 2001.

N(n) represents heart age at baseline and change in heart age after one year of follow-up.

Vascular age reported of subgroup aged 30–70 years and in primary prevention. Of the entire cohort, 42% was female.

Table 3.

Details of studies using additional tests to determine vascular age in (patient) populations.

First author	Time of data-collection	Country	Study size (% women)	Population	Vascular age approach/ test used	Additional test used	Mean chronological age	Mean vascular age	Reclassification (when chronological age is substituted by vascular age in prediction model.
First author	Time of data-collection	Country	Study size (% women)	Population	Vascular age approach/ test used	Additional test used	Mean chronological age	Mean vascular age	Up	Down
Guaraldi²¹	2006–2007	Italy	400 (26)	HIV-patients	Value-based	CAC	48	NR	NR	NR
Shaw²²	NR	USA	10377 (40)	General	Value-based	CAC	NR	NR	NR	NR
Naqvi²³	2002–2006	USA	126 (46)	General	Value-based	CAC and CIMT	57	CIMT: 62, CAC: 58	CIMT: 21%, CAC: 13%	CIMT: 13%, CAC: 15%
Khalil²⁴	NR	USA	20 (40)	General	Value-based	CAC and CIMT	56	CIMT: 69, CAC: 67	CIMT: 31%, CAC: 38%	NR
Gepner²⁵	2001–2004	USA	506 (43)	General	Value-based	CIMT	55	62	20%	9%^a
Junyent²⁶	1998–2006	Spain	409 (41)	Primary lipid disorder/ healthy controls	Value-based	CIMT	49	63	10%	13%
Kabra²⁷	NR	USA	26 (33)	Patients with STEMI	Value-based	CIMT	53	80	50%	4%
Loboz-Rudnicka²⁸	NR	Poland	187 (54)	Asymptomatic, with risk factors	Value-based	CIMT	54	60	12%	0%
Stein²⁹	2001–2002	USA	82 (45)	General	Value-based	CIMT	56	66	15%	4%
Reece^b,30	2006–2012	Australia	576 controls, 687 ODP	Opioid dependent patients and healthy controls	Value-based	PWV	Patient: 35, control: 30	Patient: 39, control: 33	NR	NR
Romanens³¹	NR	Switzerland	1500 (43)	Self-referred and physician-referred subjects free from CVD	Value-based	TPA	Derivation: 59, validation: 50	Derivation: 49, validation: 42	NR	NR
Adolphe³²	2000–2001	USA	2291 (58)	General	Risk- and value-based	CIMT	Men: 43, women: 42	Men: 39 (52), women: 38 (47)^c	NR	NR
Kawai³³	NR	USA	121/65 (100)	SLE patients/ healthy controls	Risk-based	CAC	Case: 39, control: 40^d	Case: 40, control: 36	8%	1%^e
McClelland³⁴	2000–2002	USA	6814 (53)	General	Risk-based	CAC	62	NR	12%	16%
Munir et al.³⁵	1998–2003	USA	1998 (18)	Active duty army-personnel	Risk-based	CAC	Men: 43, women: 43	Men: 44, women: 41	Men: 15%, women:<5%^f	<1%

CAC: coronary artery calcium; CIMT: carotid intima-media thickness; CVD: cardiovascular disease; HIV: human immunodeficiency virus; NR: not reported; ODP: opiate-dependent patient. PWV: pulse wave velocity; SLE: Systemic lupus erythematosus; STEMI: ST-elevation myocardial infarction; TPA: total plaque area.

Only done in 261 patients not on lipid-lowering drugs.

Multiple papers by this author on the same population were published; we have included the one with the largest study sample.

Numbers represent CIMT-derived vascular age (Framingham heart age).

Information supplied by author.

Only in patients with SLE.

Numbers not reported, calculated from graphics and percentages in text.

Table 4.

Details on studies relating vascular age to cardiovascular disease.

First author Year of publication	Year of baseline-data collection	Country	Study size (% women)	Mean follow-up (years)	Number of Events	Definition of event	Area under the receiver- operator characteristic curve of model with chronological age	Area under the receiver- operator characteristic curve of model with vascular age	Net reclassification improvement
McClelland³⁴ 2009	2000–2002	USA	6814 (53)	4	189	Incident CHD	0.76	0.83	NR
Villines^a,36 2012	1998–2003	USA	1633 (0)	5.6	8	CHD	0.74 (0.61–0.86)	0.78 (0.64–0.93)	NR
Shaw²² 2006	NR	USA	10377 (40)	5	249	All-cause mortality	0.67 (0.63–0.7)	0.75 (0.72–0.79)	NR
Romanens^b,31 2014	NR	Canada	684 (not reported)	3.3	13	Myocardial infarction	0.65	0.78	20%

CHD: coronary heart disease NR: not reported.

The cross-sectional data of this study were published by Munir et al.³⁵

These are results from a validation cohort.

The use of vascular age in patient communication

Twenty papers^{6–20,37–41} proposed using vascular age as a means of communicating cardiovascular disease risk to patients. Of these 20 papers, five described methods of calculating vascular age^37,38,40,41 and 15^6–20 applied these methods to populations. Vascular age was calculated by estimating the patient’s cardiovascular risk using an existing risk prediction model such as Framingham or Systematic Coronary Risk Evaluation (SCORE). This estimated risk score was matched with the age at which the risk was equivalent but all other risk factors were at ideal levels. For example, a 50-year old man who smokes, has a systolic blood pressure of 140 mm Hg, a High-density lipoprotein (HDL) cholesterol of 45 mg/dl and a total cholesterol levels of 200 mg/dl has an estimated cardiovascular risk of 19.3% according the Framingham risk score. The vascular age of this man would be 69, as this is the age at which a man with ideal risk factor levels (e.g. not smoking, a systolic blood pressure of 120 mm Hg and normal cholesterol levels) would also have an estimated cardiovascular risk of 19.3%. An illustration of translating cardiovascular risk scores to vascular age is presented in Figure 2.

Figure 2.

Example of deriving vascular age from risk tables.

Various ways of translating risk models to vascular age have been proposed, but Framingham heart age³⁹ was the most commonly used method. Heart age exceeded the chronological age of study participants in all studies that applied heart age in a clinical setting, with the difference in means ranging from 2–26. A point-system derived by a similar approach to the Framingham heart age, but calibrated to the Japanese population was published by Yatsuya et al.³⁷

The effect on metabolic parameters of communication of the Framingham REGICOR-risk (REgistre GIroní del COR) or heart age on top of conventional medical advice alone was examined in a randomised controlled trial among 3153 public sector workers. After 12 months of follow-up, levels of metabolic parameters had improved significantly in both risk communication groups, but most profoundly in the heart age group. Furthermore, heart age had increased in the control group and decreased in the intervention groups.¹⁴

Vascular age to improve cardiovascular risk prediction

Nineteen studies proposed to incorporate results of additional tests such as non-invasive imaging of atherosclerosis into the vascular age estimate, possibly to improve the performance of existing cardiovascular risk prediction models.^{21–36,42–44} Of these 19, three papers reported on methods^42–44 and 16 applied these methods to populations.^{21–24,26–36}

Eight studies used carotid intima-media thickness (CIMT)^23–29,32 as an additional test, eight studies used coronary arterial calcification (CAC),^{21–23,24,33–36} one study used total plaque area (TPA)³¹ and one study used pulse wave velocity (PWV).³⁰ The results from these additional tests were subsequently used to calculate the vascular age using either a value-based approach or a cardiovascular risk-based approach (Figure 3).

Figure 3.

Example of deriving vascular age from an imaging test.

Value-based approach to calculate vascular age

In the 12 studies that used a value-based approach,^21–32 vascular age was the age at which a participant’s result from an imaging test would equal the population reference values.

Stein and colleagues were the first to develop the value-based approach.²⁹ Using data from the ARIC study, they defined vascular age as the age at which the composite CIMT-value for an individual of a given race and sex would represent the median CIMT value in the ARIC study. A person with an above-average CIMT would be considered older, while a person with a below-average CIMT is considered younger. This vascular age can subsequently be used as communication tool or as substitute for chronological age in risk prediction models. A number of studies have applied this concept in various clinical settings (Table 2).

Mean vascular age was higher than chronological age in all but one study, with the differences ranging from 1–26.5 years. In the six cross-sectional studies that reported on risk reclassification using CIMT-derived vascular age, 10–51% of patients would have been classified to a higher risk category on the basis of their CIMT whereas 0–13% of the individuals were classified to a lower risk category. Two prospective studies showed that the ability of a prediction model to assign a higher predicted risk to future cases than to non-cases improved when chronological age was replaced by a test-derived vascular age in the Framingham risk score.^22,31

Risk-based approach to calculate vascular age

We identified four^33–36 studies that used a risk-based approach to define vascular age (Table 3). In the risk-based approach, vascular age is the age at which the risk for CVD equals the risk associated with the imaging-derived degree of subclinical atherosclerosis. Longitudinal data are required to calculate a risk-based vascular age; reflecting the intrinsic need for estimates on the association between age and future CVD risk, and atherosclerosis and future CVD risk.

McClelland et al.³⁴ were the first to propose and a risk-based approach in participants from the MESA study, they showed that the Receiver operating characteristic-Area Under the Curve (ROC-AUC) of the Framingham risk score increased from 0.76 in a model with chronological age to 0.83 in a model with CAC-derived vascular age. The calculator to estimate vascular age using the risk-based approach has been published on http://www.mesa-nhlbi.org/Calcium/ArterialAge.aspx, and has been used in two subsequent studies. Risk-based vascular age of patients with systemic lupus erythematosus was similar to their chronological age (40 vs 39 years) whereas the vascular age of healthy controls was 4 years lower than their chronological age (36 vs 40 years).³³ Vascular age was 44 in men and 41 in women among 1988 individuals, leading to upclassification in 15% of men and less than 5% of women, and to down classification in 1% of men.³⁵ Substitution of chronological age by CAC-derived vascular age in the Framingham risk score in men from this cohort increased the ROC-AUC from 0.74 (0.61–0.86) to 0.78 (0.64–0.93).³⁶

Discussion

The present systematic review on vascular age shows that studies have proposed different strategies to use vascular age as tool to communicate cardiovascular risk to patients or as a means to improve the performance of cardiovascular risk prediction models. Direct comparisons between different approaches to compute vascular age are currently lacking. Direct comparisons between different approaches to compute vascular age are currently lacking and data on its effectiveness in improving patient outcomes is scarce.

Vascular age in risk communication

Several studies have assessed whether vascular age as a means of communicating cardiovascular risk levels to patients leads to a better understanding of the level of risk and to improved adherence to therapies and lifestyle changes.^6,14,15 A recent trial showed that informing patients about the Framingham REGICOR risk, or heart age, on top of conventional medical advice alone significantly improved metabolic parameters in both risk communication groups, but most profoundly in the heart age group.¹⁴ Similarly, Soureti et al. compared the impact on participants' risk perceptions and intention to make lifestyle changes of communicating CVD risk in either the traditional format of a 10-year CVD percentage risk or as a heart age.³ Heart age had a greater emotional impact in younger individuals at high CVD risk levels than use of a percentage CVD risk score and would therefore be more likely to promote healthier lifestyles. Furthermore, in a study to investigate patient experiences and understanding of online heart age calculators, despite misinterpretation of risk factor questions, the calculators prompted participants to consider lifestyle changes, irrespective of their heart age level.⁶ Other studies have shown that the way physicians communicate the risks and benefits of medical care may influence patients' choices regarding treatment^45,46 and their willingness to accept treatment.⁴⁷ Moreover, patient knowledge of their coronary risk profile has shown to slightly improve the efficacy of lipid therapy.⁴

Lifetime risk has been suggested as an alternative way to present cardiovascular risk and to better reflect the long-term effects of exposure to risk factors than 10-year risk. Lifetime risk is fairly constant over a lifetime and may therefore be valuable in risk communication, especially in young- and middle aged individuals for whom 10-year risk is inherently low because of their age. Lifetime risk, however, may be less intuitive in risk communication than vascular age because lifetime risk is presented as a percentage and vascular age as an age.⁴⁸

A concept similar to vascular age, ‘lung age', has been used for smokers, when smokers were informed about their estimated lung age, smoking cessation was 7.2% higher than for patients who were not presented a lung age.⁴⁹ On the contrary, communication of the results of carotid imaging tests among smokers who were being motivated to quit smoking did not result in higher smoking cessation rates,⁵⁰ suggesting that more than a simple objective measurement is needed to motivate patients to improve their health behaviours and help them better understand their cardiovascular risk.

Vascular age in cardiovascular risk prediction

Cardiovascular risk prediction performs modestly at best and there are continuing efforts to improve it.^51,52 In the present review, we have identified several studies that have proposed the use test-derived vascular age to achieve this goal. These studies have suggested replacing chronological age by vascular age under the premise that this better reflects an individual’s exposure to cardiovascular risk factors.

Most studies added the results of an imaging-based degree of atherosclerosis, primarily CIMT and CAC, to a standard cardiovascular prediction model, without substitution of age.⁵³ While CIMT was recently shown to have modest value when added to the Framingham risk score,⁵⁴ CAC has been shown to have provide significant improvements in cardiovascular risk prediction. However, a limitation for this approach is the need for individuals to have CAC scores greater than zero. In this review, risk-based vascular age was only calculated using the results of CAC tests and was shown to materially improve the discriminative ability of the risk prediction model.^34,36 Whether such improvements are achievable using CIMT-derived risk-based vascular age needs to be determined. However, based on the results of previous studies where CIMT and CAC were added to a risk prediction model,^53,54 it might be expected that predictive accuracy is better improved by substituting a CAC-derived vascular age to chronological age rather than using a CIMT-derived vascular age.

Next to the choice upon which measure should be used to base vascular age, the potential added value of vascular age in cardiovascular risk prediction is likely to depend on the way vascular age will be defined as well. We identified two distinct methods to define a test-derived measure of vascular age, i.e. a value-based approach and a risk-based approach. In the value-based approach, vascular age is based on the expected age of an individual based on the reference values derived from a healthy population. In the risk-based approach, vascular age represents the age where the cardiovascular risk estimated by imaging a patient’s atherosclerotic burden or another test result equals the calculated CVD risk based on age. Therefore, in a value-based approach vascular age is a direct translation of the result of a test, whereas in risk-based vascular age information is added.

Intuitively, the risk-based approach might make more biological sense since it matches the chronological age equivalent to the risk associated with a given level of atherosclerosis, and not to the age where an average person would be expected to have a certain level of atherosclerosis (i.e. the reference value-based approach). However, studies comparing these two different approaches are lacking and would provide further insights in the value of vascular age assessment in patient communication and for CV risk prediction.

Strengths and limitations

To our knowledge, this is the first systematic review on the concepts behind vascular age. Such a review is needed since the expression ‘vascular age' is increasingly used, yet the exact definition and the potential uses in a clinical setting remain ambiguous. A limitation of this paper is that the search and screening process were conducted by one author (KAG). This could have increased the possibility of missing some significant papers. However, we cross-checked the references listed in the selected papers and included relevant studies that were not identified in our initial search. Moreover, since the aim of our study was to provide an overview of the current concepts behind vascular age rather than to compare the different methods and evaluate their clinical application, definite conclusions about the value of vascular age are yet to be reached.

Conclusion

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflicts of interest

None declared.

References

World Health Organization (WHO). Global status report on noncommunicable diseases 2011. World Health Organization. http://www.who.int/nmh/publications/ ncd_report2010/en/.

Wilson

PWF

D’Agostino

Levy

. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97: 1837–1847.

Soureti

Hurling

Murray

. Evaluation of a cardiovascular disease risk assessment tool for the promotion of healthier lifestyles. Eur J Cardiovasc Prev Rehabil 2010; 17: 519–523.

Grover

Lowensteyn

Joseph

. Patient knowledge of coronary risk profile improves the effectiveness of dyslipidemia therapy: The check-up study: A randomized controlled trial. Arch Intern Med 2007; 167: 2296–2303.

Anderson

Grégoire

Hegele

. 2012 Update of the Canadian Cardiovascular Society guidelines for the diagnosis and treatment of dyslipidemia for the prevention of cardiovascular disease in the adult. Can J Cardiol 2013; 29: 151–167.

Bonner

Jansen

Newell

. I don’t believe it, but I'd better do something about it: Patient experiences of online heart age risk calculators. J Med Internet Res 2014; 16: e120–e120.

Daher

Daigle

Bhatia

. The prevention of cardiovascular disease in cancer survivors. Texas Hear Inst J 2012; 39: 190–198.

Davies

Gompels

Johnston

. Mind the gap: Difference between Framingham heart age and real age increases with age in HIV-positive individuals – a clinical cohort study. BMJ Open 2013; 3: e003245–e003245. DOI: 10.1136/bmjopen-2013-003245.

De Socio

GVL

Martinelli

Ricci

. Relations between cardiovascular risk estimates and subclinical atherosclerosis in naive HIV patients: Results from the HERMES study. Int J STD AIDS 2010; 21: 267–272.

10.

García-Ortiz

Recio-Rodríguez

Schmidt-Trucksäss

. Relationship between objectively measured physical activity and cardiovascular aging in the general population–the EVIDENT trial. Atherosclerosis 2014; 233: 434–440.

11.

Hollar DW and Lewis JS. Heart age differentials and general cardiovascular risk profiles for persons with varying disabilities: NHANES 2001–2010. Disabil Health J 2015; 8: 51--60.

12.

Iancu

Copăescu

Şerban

. Laparoscopic sleeve gastrectomy reduces the predicted coronary heart disease risk and the vascular age in obese subjects. Chirurgia (Bucur) 2013; 108(5): 659–665.

13.

Kakinami

Block

Adams

. Risk of cardiovascular disease in HIV, hepatitis C, or HIV/hepatitis C patients compared to the general population. Int J Clin Pract 2013; 67: 6–13.

14.

Lopez-Gonzalez AA, Aguilo A, Frontera M, et al. Effectiveness of the heart age tool for improving modifiable cardiovascular risk factors in a Southern European population: A randomized trial. Eur J Prev Cardiol 2014. [Epub ahead of print: 3 February 2014]. DOI:10.1177/2047487313518479.

15.

Neufingerl

Cobain

Newson

. Web-based self-assessment health tools: Who are the users and what is the impact of missing input information? J Med Internet Res 2014; 16: e215–e215.

16.

Ratliff

Palmese

Reutenauer

. Obese schizophrenia spectrum patients have significantly higher 10-year general cardiovascular risk and vascular ages than obese individuals without severe mental illness. Psychosomatics 2013; 54: 67–73.

17.

Sharma

Sahoo

Shah

. Are Gujarati Asian Indians ‘older' for their ‘vascular age' as compared to their ‘chronological age'? QJM 2014; 113. DOI: 10.1093/qjmed/hcu158. [Epub ahead of print 1 August 2014].

18.

Soares

Piovesan

Gustavo

A da S

. Alimentary habits, physical activity, and Framingham global risk score in metabolic syndrome. Arq Bras Cardiol 2014; 102: 374–382.

19.

Tay

Nurjono

Lee

. Increased Framingham 10-year CVD risk in Chinese patients with schizophrenia. Schizophr Res 2013; 147: 187–192.

20.

Zhang

Lelong

Kretz

. Characteristics and future cardiovascular risk of patients with not-at-goal hypertension in general practice in France: The AVANT’AGE study. J Clin Hypertens 2013; 15: 291–295.

21.

Guaraldi

Zona

Alexopoulos

. Coronary aging in HIV-infected patients. Clin Infect Dis 2009; 49: 1756–1762.

22.

Shaw

Raggi

Berman

. Coronary artery calcium as a measure of biologic age. Atherosclerosis 2006; 188: 112–119.

23.

Naqvi

Mendoza

Rafii

. High prevalence of ultrasound detected carotid atherosclerosis in subjects with low Framingham risk score: Potential implications for screening for subclinical atherosclerosis. J Am Soc Echocardiogr 2010; 23: 809–815.

24.

Khalil

Mukete

Durkin

. A comparison of assessment of coronary calcium vs carotid intima media thickness for determination of vascular age and adjustment of the Framingham risk score. Prev Cardiol 2010; 13: 117–121.

25.

Gepner

Keevil

Wyman

. Use of carotid intima-media thickness and vascular age to modify cardiovascular risk prediction. J Am Soc Echocardiogr 2006; 19: 1170–1174.

26.

Junyent

Zambón

Gilabert

. Carotid atherosclerosis and vascular age in the assessment of coronary heart disease risk beyond the Framingham Risk Score. Atherosclerosis 2008; 196(2): 803–809.

27.

Kabra

Neri

Weiner

. Carotid intima-media thickness assessment in refinement of the Framingham risk score: Can it predict ST-elevation myocardial infarction? A pilot study. Echocardiography 2013; 30: 1209–1213.

28.

Łoboz-Rudnicka

Jaroch

Bociąga

. Relationship between vascular age and classic cardiovascular risk factors and arterial stiffness. Cardiol J 2013; 20: 394–401.

29.

Stein

Fraizer

Aeschlimann

. Vascular age: Integrating carotid intima-media thickness measurements with global coronary risk assessment. Clin Cardiol 2004; 27: 388–392.

30.

Reece

Hulse

. Impact of lifetime opioid exposure on arterial stiffness and vascular age: Cross-sectional and longitudinal studies in men and women. BMJ Open 2014; 4: e004521–e004521.

31.

Romanens

Ackermann

Sudano

. Arterial age as a substitute for chronological age in the AGLA risk function could improve coronary risk prediction. Swiss Med Wkly 2014; 144: w13967–w13967.

32.

Adolphe

Huang

Cook

. Carotid intima-media thickness determined vascular age and the Framingham risk score. Crit Pathw Cardiol 2011; 10: 173–179.

33.

Kawai

Solus

Oeser

. Novel cardiovascular risk prediction models in patients with systemic lupus erythematosus. Lupus 2011; 20: 1526–1534.

34.

McClelland

Nasir

Budoff

. Arterial age as a function of coronary artery calcium (from the Multi-Ethnic Study of Atherosclerosis (MESA)). Am J Cardiol 2009; 103: 59–63.

35.

Munir

Bauer

. Impact of coronary calcium on arterial age and coronary heart disease risk estimation using the MESA arterial age calculator. Atherosclerosis 2010; 211: 467–470.

36.

Villines

Taylor

. Multi-ethnic study of atherosclerosis arterial age versus Framingham 10-year or lifetime cardiovascular risk. Am J Cardiol 2012; 110: 1627–1630.

37.

Yatsuya

Iso

Yamagishi

. Development of a point-based prediction model for the incidence of total stroke: Japan public health center study. Stroke 2013; 44: 1295–1302.

38.

Marma

Lloyd-Jones

. Systematic examination of the updated Framingham Heart Study general cardiovascular risk profile. Circulation 2009; 120: 384–390.

39.

D’Agostino

Vasan

Pencina

. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation 2008; 117(6): 743–753.

40.

Cuende

Calaveras-Lagartos

. How to calculate vascular age with the SCORE project scales: A new method of cardiovascular risk evaluation. Eur Heart J 2010; 31: 2351–2358.

41.

Cooney

Vartiainen

Laatikainen

. Cardiovascular risk age: Concepts and practicalities. Heart 2012; 98: 941–946.

42.

Ball

Feiveson

Schlegel

. Predicting ‘heart age' using electrocardiography. J Pers Med 2014; 4: 65–78.

43.

Schisterman

Whitcomb

. Coronary age as a risk factor in the modified Framingham risk score. BMC Med Imaging 2004; 4: 1–1.

44.

Sirineni

GKR

Raggi

Shaw

. Calculation of coronary age using calcium scores in multiple ethnicities. Int J Cardiovasc Imaging 2008; 24: 107–111.

45.

Man-Son-Hing

O’Connor

Drake

. The effect of qualitative vs. quantitative presentation of probability estimates on patient decision-making: A randomized trial. Heal Expect 2002; 5: 246–255.

46.

Halvorsen

Selmer

Kristiansen

. Different ways to describe the benefits of risk-reducing treatments: A randomized trial. Ann Intern Med 2007; 146: 848–856.

47.

Misselbrook

Armstrong

. Patients’ responses to risk information about the benefits of treating hypertension. Br J Gen Pract 2001; 51: 276–279.

48.

Wilkins

Ning

Berry

. Lifetime risk and years lived free of total cardiovascular disease. JAMA 2012; 308: 1795–1801.

49.

Parkes

Greenhalgh

Griffin

. Effect on smoking quit rate of telling patients their lung age: The Step2quit randomised controlled trial. Br Med J 2008; 336: 598–600.

50.

Rodondi

Collet

T-H

Nanchen

. Impact of carotid plaque screening on smoking cessation and other cardiovascular risk factors: A randomized controlled trial. Arch Intern Med 2012; 172: 344–352.

51.

Siontis

GCM

Tzoulaki

Siontis

. Comparisons of established risk prediction models for cardiovascular disease: Systematic review. Br Med J 2012; 344: e3318–e3318.

52.

Folsom

. Classical and novel biomarkers for cardiovascular risk prediction in the United States. J Epidemiol 2013; 23: 158–162.

53.

Peters

SAE

den Ruijter

Bots

. Improvements in risk stratification for the occurrence of cardiovascular disease by imaging subclinical atherosclerosis: A systematic review. Heart 2012; 98: 177–184.

54.

den Ruijter

Peters

SAE

Anderson

. Common carotid intima-media thickness measurements in cardiovascular risk prediction: A meta-analysis. JAMA 2012; 308: 796–803.