Intern Emerg Med (2011) 6 (Suppl 1):S45S51

DOI 10.1007/s11739-011-0669-5

SYMPOSIUM WHAT INTERVENTION TRIALS DONT TELL US: THE RESIDUAL RISK

The epidemiological concept of residual risk

Diego Vanuzzo

 SIMI 2011

Abstract Residual cardiovascular risk can be defined as

the residual risk of incident vascular events or progression
of established vascular damage persisting in patients treated with current evidence-based recommended care
including the risk that established from risk factors, such as
dyslipidemia, high blood pressure, and the risk related to
emerging or newer risk factors. The concept clearly derives
from intervention trials, mainly the statin trials, and there is
a lot of debate about the residual risk conferred by other
lipid components, in particular low levels of HDL cholesterol and high levels of triglycerides. A meta-analysis of
53 fibrates (16,802 subjects) and 30 niacin trials (4,749
subjects) revealed an average HDL-C increase of 10% with
fibrates and 16% with niacin, a triglyceride decrease of
36% with fibrates and 20% with niacin, and a LDL-C
decrease of 8% with fibrates and 14% with niacin. These
lipid changes resulted in similar overall reductions in major
coronary events evidenced by a 25% decrease with fibrates
and 27% with niacin. However, recent analyses of the
primary and secondary prevention trials like JUPITER,
Treating to New Targets (TNT) and PROVE-IT TIMI 22
force to reconsider the issue. In these three trials, HDL-C
was useful in the initial risk assessment but when LDL-C
was aggressively lowered the residual risk predictive value
of HDL-C was markedly attenuated. Also epidemiological
studies evaluate the residual risk in treated hypertensives
and dyslipidemic subjects within a general population. The
PRIME study in Northern Ireland and France and the
Progetto CUORE study in Italy, both with a 10-year

D. Vanuzzo (&)
Cardiovascular Prevention Centre, Health Unit 4 Medio
Friuli, P.le Santa Maria Misericordia, 33100 Udine, Italy
e-mail: diego.vanuzzo@mediofriuli.it

follow-up were able to test the hypothesis of the residual

cardiovascular risk in treated hypertensives, because the
proportion of treated dyslipidemic subjects was too low at
baseline. In both studies treatment with antihypertensive
agents was associated with a sizeable residual cardiovascular risk with the hazard ratio of 1.51.7, suggesting that
more efficient risk reduction strategies in hypertension
should be developed as a priority. In conclusion residual
cardiovascular risk should be better studied in cardiovascular epidemiology, refining the methods to evaluate it, to
consider measures of exposure to the modifiable risk factors and indicators of treatment (both at pharmacological
and lifestyle level) over the time. Repeated measures and
cohortal follow-up are needed and also new statistical
methods are necessary to evaluate the residual risk to
understand how to reduce it.
Keywords Cardiovascular residual risk
Lipid trials

Longitudinal epidemiological studies

Residual risk of treated hypertensives

Recently, Hermans and Fruchart proposed the following

holistic definition of residual cardiovascular risk (RCVR):
the residual risk of incident vascular events or progression
of established vascular damage persisting in patients treated with current evidence-based recommended care,
including risk from established risk factors, such as dyslipidemia, high blood pressure, hyperglycemia, inflammation and unhealthy lifestyles, and risk related to emerging
or newer risk factors [1]. These authors comment that,
while evidence-based current standards of care de facto
represent a paradigmatic frame, RCVR is a dynamic, everevolving concept that develops in close relation with
progressive upgrades, revisions or shifts in guidelines,

123

S46

recommendations and state-of-the-art consensuses defining

standards of care [1]. The concept clearly derives from
interventional trials, mainly the statin trials, and there is a
lot of debate about the residual risk conferred by other lipid
components, in particular low levels of HDL cholesterol
and high levels of triglycerides. Randomized clinical trials
are cited routinely as the highest form of clinical epidemiology [2], and clinical epidemiology is the science of
making predictions about individual patientsusing strong
scientific methods to obtain the kind of information
clinicians need to make good decisions in the care of
patients. [3]. Recent interest has focused also on the
ability of observational methods to yield valuable insights
into the nature and effect of treatments of common and
serious diseases that represent major threats to the publics
health [4], like cardiovascular diseases. Management of
individual patients requires the integration of the results of
both observational and interventional epidemiology. Also
in the field of hypertension, observational epidemiology
clearly demonstrates that hypertensive patients have an
increased risk of developing cardiovascular complications.
[5]. In addition, blood pressure decrease is associated with
a reduction in major cardiovascular complications: coronary heart disease (CHD), stroke, congestive heart failure
and cardiovascular mortality [6]. To improve hypertensive
treatment, we should remember that although antihypertensive treatment reduces cardiovascular risk, it cannot
completely reverse the hypertension-induced risk of morbid events. Therefore, antihypertensive treatment cannot be
considered effective if only part of the cardiovascular
events attributable to hypertension is prevented. Conventional therapeutic trials do not provide this information
because patients are highly selected. Other sources of
evidence are needed to determine the proportion of
hypertension-induced cardiovascular events prevented by
antihypertensive drugs, as they are prescribed and monitored in the general population. Little is known about the
cardiovascular risk of patients receiving antihypertensive
treatment, after the levels of risk factors achieved, that is
their RCVR, is taken into account. Some data support a
residual cardiovascular risk in treated hypertensives, even
in patients with an adequate control of their blood pressure
levels [7, 8]. Residual cardiovascular risk is important at
the public health level to implement new guidelines, and at
the epidemiological level to build more realistic prediction
models, taking into account both the current level of the
risk factor and the type of drug treatment. The confirmation
of an important residual cardiovascular risk in hypertensive
patients or in hypercholesterolaemic patients has important
clinical implications, opening discussions on earlier treatment, more intensive treatment and the need for developing
new drugs [9]. In this review, we will consider the concept
of RCVR in lipid lowering trials and in observational

123

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

epidemiology of treated risk conditions like hypertension

and dyslipidemia.

Residual cardiovascular risk in lipid lowering trials

Impacting on cardiovascular disease (CVD) through identification and reduction of exposure to risk factors (RFs)
has met with substantial success over the past decades,
especially in the field of hypercholesterolemia. Given the
continuous progression of the obesity pandemic and
worldwide rising rates of metabolic syndrome, identifying
and impacting upon RCVR in patients benefiting from
current, evidence-based standards of care is increasingly
becoming an unmet need in CV risk reduction [1, 10, 11].
Current guidelines for standards of care emphasize the
importance of multifactorial intervention to achieve
recommended levels of LDL-C, blood pressure and, in
diabetic patients, glycemic control [1214]. The importance of determining RCVR in high-risk populations lies in
the fact that:

A substantial fraction of RCVR remains modifiable,

A major component of RCVR is lipid- or hypertensionrelated,
RCVR is amenable to reduction through various
lifestyle or pharmacologic interventions [1].

Management guidelines for dyslipidemia, including

those for diabetes, have primarily focused on reducing
hypercholesterolemia (i.e., the level of cholesterol associated with major circulating atherogenic lipoproteins, the
apoB100-containing lipoproteins [mostly LDL] and/or total
cholesterol [1214]. Statin therapy is the mainstay of
dyslipidemia management, and statin efficacy in primary
or secondary CV prevention is supported by convergent
evidence from large prospective, randomized clinical
trials.[1521] All major guidelines recommend aggressive
treatment of LDL-C in patients with hypercholesterolemia
in primary or secondary CV prevention. In practice, control
of LDL-C usually means intervention with a statin as the
preferred first agent when lifestyle interventions (e.g., diet
and exercise) are insufficiently effective. Despite clear
guidelines, many patients on statins often do not attain their
respective targets. Relative cardiovascular risk reduction
after statin therapy is approximately 2535%, depending
on the baseline level of risk. Additional lowering of LDL-C
with high-dose statins or the administration of ezetimibe,
while allowing for further CV risk reduction and target
attainment, leaves out a considerable lipid-related RCVR
due to abnormal non-LDL lipid constituents, unaddressed
or little improved by statins[1]. A considerable part of poststatin lipid abnormalities associated with RCVR appear to
result from abnormal levels of triglyceride (TG)-rich

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

lipoproteins, of their remnants and from a decreased

number and/or functionality of HDLs.[1] These non-LDL
lipoproteins are the main effectors of TG transport from
liver or gut to TG-consuming organs and of reverse
cholesterol transport (RCT). The most prevalent form of
non-LDL dyslipidemia is known as atherogenic dyslipidemia or atherogenic dyslipoproteinemia. The distinguishing features of atherogenic dyslipidemia are a
decrease in HDL-C levels together with raised fasting TG
levels. Atherogenic dyslipidemia is commonly observed
among patients treated for high LDL-C, and is an important
effector of absolute and relative CV risk, even in the setting
of LDL-C at target. While strictly speaking raised LDL-C
levels also represent a state of dyslipidemia predisposing to
atherosclerosis, this abnormality should be better defined as
hypercholesterolemia. The recent results of the Arterial
Biology for the Investigation of the Treatment Effects of
Reducing Cholesterol-6 HDL and LDL Treatment Strategies (ARBITER-6 HALTS) and the Action to Control
Cardiovascular Risk in Diabetes (ACCORD) have reinvigorated the concept of combining other lipid modifying
agents to ongoing statin therapy [22, 23]. In ARBITER-6
HALTS, two statin combination groups were compared:
one randomized to statin plus ezetimibe with an achieved
LDL-C mean level of 66 mg/dl while the other achieved an
LDL-C 70 mg/dl with statin plus niacin [22]. In ACCORD,
the mean LDL-C achieved was 81 mg/dl on the combination statin plus fenofibrate [23]. The ARBITER-6 HALTS
and ACCORD trials both enrolled high risk patients in
whom 3637% had the combination of diabetes and documented CHD. ARBITER-6 HALTS was a smaller study
of 363 subjects, with only 208 of them completing the trial,
resulting in only 2 CV events in 14 months (1% event rate)
with statin-niacin and 9 events with statin to ezetimibe
ratio (5% event rate). ACCORDs 10-year projected CV
event rate (first occurrence of nonfatal myocardial infarction, nonfatal stroke, or death from cardiovascular causes)
was 22% with statin to fenofibrate ratio compared to 24%
on statin to placebo ratio. Does the residual risk in these
trials reflect a potential strategy based on a more aggressive
approach to lipoproteins other than LDL-C? A recent
hospital-based case-controlled study of incident CHD in
patients presenting with LDL-C \130 mg/dl (mean baseline LDL-C 73 mg/dl) suggested that high triglyceride and
low HDL-C strongly and synergistically contributed to
CHD [3]. Although 61% of patients were on statins, both
low HDL-C and high triglyceride levels were still associated with CHD events even when LDL-C levels were
\70 mg/dl. In this analysis, the odds of CHD events
increased by 20% per 23 mg/dl increase in TG and by 40%
per 7.5 mg/dl decrease in HDL-C [24]. In both the
ARBITER-6 HALTS and ACCORD trials, there appeared
to be continued benefit in those subjects treated with either

S47

niacin (ARBITER-6) or fenofibrate (ACCORD) who were

at the lowest baseline LDL-C. In ARBITER-6, the ratio of
statin to niacin combination resulted in a significantly
greater efficacy on carotid intima-media thickness
(p = 0.003) as well as improved CV events (p = 0.04)
when compared to statin:ezetimibe, despite greater LDL-C
lowering with the ratio of statin to ezetimibe (8466 mg/dl)
compared to the ratio of statin to niacin (8070 mg/dl). It
should be noted that the ratio of statin to niacin significantly raised HDL-C from 42.5 to 50 mg/dl and lowered
triglycerides from 122 to 103 mg/dl whereas the ratio of
statin to ezetimibe lowered HDL-C from 43 to 42 mg/dl
and had no effect on the triglyceride level of 126 mg/dl. In
ACCORD, approximately 60% of patients were on baseline statin therapy and fenofibrate combined with simvastatin raised HDL-C from 38 to 41.2 mg/dl while lowering
TG from 162 to 122 mg/dl. The lowest LDL-C tertile
group in ACCORD had baseline LDL-C levels \84 mg/dl
and trended toward benefit with a 23% reduction in CV
events in those treated with fenofibrate compared to
placebo. In addition, those at the highest TG tertile
of [204 mg/dl and lowest HDL-C tertile of \34 mg/dl
demonstrated a 31% decrease in CV events (p = 0.057).
Although these results are merely hypothesis generating,
they do suggest that once LDL-C approaches the goal of
70 mg/dl, medications such as niacin and fenofibrate targeting low HDL and/or high TG may exhibit additional
benefits in reducing the residual risk. A meta-analysis of 53
fibrate (16,802 subjects) and 30 niacin trials (4,749 subjects) performed before either ARBITER-6 HALTS or
ACCORD trials were published, revealed an average HDL-C
increase of 10% with fibrates and 16% with niacin, a triglyceride decrease of 36% with fibrates and 20% with niacin,
and a LDL-C decrease of 8% with fibrates and 14% with
niacin. These lipid changes resulted in similar overall
reductions in major coronary events evidenced by a 25%
decrease with fibrates and 27% with niacin [25]. However, a
recent analysis of the primary prevention JUPITER trial
forces to reconsider the issue [26]. In the 8,901 JUPITER
subjects randomized to placebo with a mean LDL-C 100 mg/
dl, HDL-C levels were inversely related to vascular events.
Conversely, the 8,900 subjects randomized to rosuvastatin
20 mg and treated to a mean LDL-C of 54 mg/dl demonstrated no significant relationship between HDL-C and
vascular risk [26]. As in JUPITER, in both treating to new
targets (TNT) and PROVE-IT TIMI 22 trials, HDL-C
was useful in the initial risk assessment, but when LDL-C
was aggressively lowered by atorvastatin 80 mg to 77 mg/dl
and to 62 mg/dl respectively, the residual risk predictive
value was markedly attenuated [27, 28]. In TNT, the baseline
HDL-C 47 mg/dl was not changed by atorvastatin 80 mg, in
PROVE-IT the baseline HDL-C 39 mg/dl showed an
increase of 6.5% (2.5 mg/dl), whereas in JUPITER a 4%

123

S48

increase (2 mg/dl) was observed. Of interest, in the TNT

study those who achieved a dual goal of LDL-C\70 mg/dl
and HDL [55 mg/dl exhibited a 40% reduction in major
adverse cardiac events (MACE) and when the entire cohort
was evaluated, rather than just those allocated to atorvastatin 80 mg, HDL-C was predictive of MACE even when
the LDL-C levels were \70 mg/dl [27]. Similarly, in a
post-hoc analysis of 4 secondary prevention randomized
statin trials, atherosclerotic regression occurred only in
those achieving both LDL-C \87.5 mg/dl and an increase
in HDL-C of [7.5% [29]. This concept was further
endorsed in a post-hoc analysis of a primary prevention
population in AFCAPS/TexCAPS where those receiving
the greatest reductions in CV events exhibited both LDL-C
reductions [25% and HDL-C increases [7.5% [30].
As also noted in AFCAPS/TexCAPS, the benefit from
lovastatin appeared to be greatest in those with baseline
HDL-C levels \40 mg/dl, thus suggesting statin therapy
neutralized the negative impact associated with a low
HDL-C [31].

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

France) [9]. In the whole cohort, a total of 796 patients

developed a fatal or non-fatal cardiovascular event. Antihypertensive drug use at baseline was significantly associated (RR = 1.50, 95% CI 1.251.80) with total
cardiovascular event risk, but not lipid-lowering drug use,
after adjusting for classic risk factors (age, smoking, total
cholesterol, high-density lipoprotein cholesterol, systolic
blood pressure and diabetes). Similar results were obtained
for coronary heart disease (RR = 1.46, 95% CI 1.181.80),
stroke (RR = 1.75, 95% CI: 1.142.70) and cardiovascular
death (RR = 1.62, 95% CI: 1.022.58), but neither for
total death (RR = 1.15, 95% CI 0.891.48) nor for noncardiovascular death (RR = 1.00, 95% CI 0.741.36). For
any cardiovascular end point, residual risks did not globally
differ according to the antihypertensive drug class prescribed at baseline. In conclusion, treatment with antihypertensive agents, but not with lipid-lowering agents, was
associated with a sizeable residual cardiovascular risk,
suggesting that more efficient risk reduction strategies in
hypertension should be developed as a priority [9]. The
lipid-lowering results may be due to the lower prevalence
at baseline compared to that of antihypertensive drugs.

Residual cardiovascular risk in observational studies

In the PRIME (Prospective Epidemiological Study of
Myocardial Infarction) study, the authors evaluated if a
10-year follow-up might give enough power to test the
presence of specific residual cardiovascular risks in treated
hypertensives, not only including CHD, but also stroke,
total of fatal and non-fatal cardiovascular events and cardiovascular mortality. Furthermore, because hypercholesterolaemia might behave similarly, they also tested the
presence of a residual cardiovascular risk associated with
lipid-lowering drugs [9]. The PRIME study was established
in 1991 in the populations of four WHO MONICA Collaborating Centres in Belfast (UK), and Lille, Strasbourg
and Toulouse (France). The target was to recruit 2,500
men, aged 5059 years, in each centre and to follow them
for over 10 years. The sample was recruited to broadly
match the social class structure of the background population [9]. All subjects receiving antihypertensive, lipidlowering or antidiabetic therapy were included in the
analysis. Antihypertensive agents (prevalence of 9.0% in
Northern Ireland and 14.3% in France) were divided into
b-blockers, diuretics, angiotensin-converting enzyme
(ACE) inhibitors, calcium channel antagonists (CCAs),
a-blockers and centrally acting agents, with the latter two
being grouped together in the analysis because of small
numbers. Subjects taking drug combinations were included
in the analysis as if they were taking the drugs separately.
Lipid-lowering agents were divided into fibrates (prevalence of 0.6% in Northern Ireland and 7.3% in France) and
statins (prevalence of 0.3% in Northern Ireland and 3.7% in

123

Final considerations
To understand the results of the intervention trials and of
the observational PRIME study, some hypotheses can be
considered.
1.

The concept of lifetime risk suggests that many

individuals deemed at moderate or even low risk over
10 years may very well have a high lifetime risk for a
CVD event [32]. This concept suggests that the risk of
continued exposure to high normal blood pressure
values or abnormal lipoproteins even at moderate
levels may be compounded over time resulting in both
progressive atherosclerosis and resultant CV events.
However, in individuals with lifetime LDL-C levels
reduced by approximately 28% (*40 mg/dl) due to a
nonsense mutation of PCSK9 gene, up to an 88%
reduction in CHD events has been reported [33]. This
88% reduction is in contradistinction to the 30%
reduction predicted by a similar LDL-C decrease noted
in a meta-analysis of statin trials [34]. The fact that
individuals with the PCSK9 mutations have lower
LDL-C levels throughout their entire lives essentially
triples the risk reduction when compared to those who
lowered their LDL-C to a similar magnitude, but
measured for only a 5 year span. In concert with
lifelong reductions in LDL-C from genetic mutations
such as PCSK9, the concept of continued statin therapy
beyond 5 years suggests a widening gap between those

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

S49

Fig. 1 Distribution of The

CUORE project cohorts,
number of persons (N) and
events for men and women aged
3569 years free of previous
cardiovasular disease at
baseline. Follow-up refers to
median duration in years

Table 1 Results of Cox regressions, best model for men and women aged 3569 years free of previous cardiovascular disease at baseline. The
CUORE project samples
Men
b

Women
p value

HR more
adverse
levela

HR more
favourable
levelb

p value

HR more
adverse
levela

HR more
favourable
levelb

Age (years)

b1

0.076

\0.0001

2.01

0.50

0.079

\0.0001

1.95

0.51

Sistolic blood pressure (mmHg)

b2

0.013

\0.0001

1.31

0.76

0.016

\0.0001

1.41

0.71

Total cholesterol (mg/dl)

b3

0.006

\0.0001

1.31

0.76

0.003

0.0362

1.14

0.88

HDL-cholesterol (mg/dl)

b4

-0.013

\0.0001

1.60

0.60

-0.015

0.0002

1.26

0.80

Smoking habit, yesno

b5

0.508

\0.0001

1.66

0.60

0.773

\0.0001

2.17

0.46

Diabetes, yesno

b6

0.462

0.0005

1.59

0.63

0.339

0.0859

1.40

0.71

Hypertension treatment, yesno

b7

0.490

\0.0001

1.63

0.61

0.590

\0.0001

1.80

0.55

G (l)

6.583

6.016

Survival at baseline, S (t)

0.953

0.989

b is Cox coefficient; HR is Hazard Ratio

a

For continuous variables, hazard ratio with level 1 standard deviation higher (see Table 1); for dichotomized variables, yes versus no

For continuous variables, hazard ratio with level 1 standard deviation lower (see Table 1); for dichotomized variables, no versus yes

G(l) is the linear combination of the risk factor averages or of the prevalence in each category for the corresponding b1 coefficients
S(t) is 10-year survival evaluated at mean value of risk factors
Risk equation is: 1[S(t)]^{EXP[b1 9 age ? b2 9 SBP ? b3 9 TC ? b4 9 HDL-C ? b5 (if smoker) ? b6 (if diabetic) ? b7 (if under
hypertension treatment)G(l)]}, where SBP is systolic blood pressure, TC is total cholesterol, HDL-C is HDL-cholesterol

treated with statins and not treated in both primary and

secondary patients [35, 36]. Over longer time spans,
such as decades, the sustained benefit from continuous
statin therapy achieving lower LDL-C levels may
approach the 88% reductions noted with lifelong
PCSK9 mutations.

2.

3.

The compliance with drug treatment is rather disappointing, even in clinical trials, and part of the residual risk can
be attributed to dose reduction or periodical interruption.
In observational studies the drug treatment is generally
recorded only at baseline and there is no specific
follow-up.

123

S50

4.

5.

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

Unhealthy lifestyles, which are difficult to monitor,

contribute to RCVR. Lifestyle changes in smoking
cessation, nutrition, and physical activity typically
show excellent cost-effectiveness in lowering the
burden of disease, especially with respect to cardiovascular diseases [37].
The risk prediction tools may consider the residual risk
conveyed by an exposure requiring drug treatment,
especially for hypertension, but they must be
improved. In Italy, within the Progetto CUORE,
we were able to incorporate hypertension drug treatment as a tool to reduce the effect of RCVR in the
predictive model. In Italy, the CUORE project, funded
by the Italian Ministry of Health, was launched at the
end of 1990s, with the aim of building a database of
cardiovascular risk factors collected within longitudinal studies started in the 1980s and 1990s using
standardized procedures and methods and implementing the follow-up of cardiovascular mortality and
morbidity for assessing cardiovascular risk in the adult
Italian population [38]. The CUORE project pooled
standardized data from random population samples in
northern, central, and southern Italy: 20,647 (7,520
men and 13,127 women) (Fig. 1) aged 3569 years
without previous cardiovascular events were included
in the analysis and followed-up until 2002 (10-year
mean period) for CVD. As women usually develop less
CVD events than men, more women than men were
involved in the study. Data collection procedures and
methods were standardized and are comparable with
those of the MONICA project [39]. Participation rates
were 6478% except for one cohort (MATISS 1987
with 40%). A detailed description of these cohorts is
available on the CUORE project web site (http://
www.cuore.iss.it/eng/default.asp). At baseline, risk
factors were assessed by standardized procedures.
Blood pressure (right arm) was measured twice with a
mercury sphygmomanometer, participant sitting, after
5 min of rest; systolic blood pressure (SBP) or diastolic blood pressure were recorded (first and fifth
phase of Korotkoff sounds); the first and second
measurements were averaged for analyses. Total cholesterol and HDL cholesterol (TC, HDL-C) were
assayed on serum in four different laboratories under
the quality control of reference laboratories and measurements were carried out using the colorimetric
enzymatic method. Information was collected by
questionnaire on cigarette smoking, personal history of
myocardial infarction, stroke, diabetes mellitus, hospitalization for cardiovascular events, and medication.
Standardized methods were used to identify and validate
suspected events. Major cardiovascular events, in particular myocardial infarction, stroke, revascularization

123

treatment (bypass or angioplasty), and sudden death,

were identified through record linkage between mortality and hospital discharge registers over December
2002. In order to identify the aetiological role of risk
factors, considered singly and in combination, and to
choose the best model, the statistical analysis was
comprised of several steps: description of risk factors
at baseline, correlation analysis among risk factors,
univariate analysis, multivariate analysis and model
effectiveness testing. Table 1 shows the coefficients of
risk factors included in the best model and hazard
ratios with level 1 standard deviation higher and lower
for continuous variables and yes versus no and no
versus yes for categorical variables. Comparison
between aetiological weight of single risk factors is
thus enabled. The RCVR attributed to hypertension
treatments, expressed as hazard ratio is 1.63 for men
and 1.80 for women, comparable to smoking and
diabetes and therefore we decided to incorporate it in
the CUORE risk score (freely available at http://www.
cuore.iss.it/sopra/calcrischio_en.asp which has also an
English version). The prevalence of lipid-lowering
drugs at baseline was too low in the Progetto CUORE
to evaluate if there was a RCVR.
In conclusion, RCVR should be taken into account also
in cardiovascular epidemiology, refining the methods to
evaluate it, in order to consider measures of exposure to the
modifiable risk factors and indicators of treatment (both at
pharmacological and lifestyle level) over the time. Repeated measures and cohortal follow-up are needed and also
new statistical methods are necessary to evaluate RCVR in
order to understand how to reduce it.
Conflict of interest

None.

References
1. Hermans MP, Fruchart JC (2010) Reducing Residual Vascular
Risk in Patients with Atherogenic Dyslipidemia: Where do we go
from here? Clin Lipidology. 5(6):811826
2. Vandenbroucke JP (2004) When are observational studies as
credible as randomised trials? Lancet 363:17281731
3. Fletcher RH, Fletcher SW, Wagner EH. Clinical Epidemiology:
The Essentials. 3rd edn. Baltimore, Md: Williams & Wilkins;
1996:2
4. Lauer MS (2001) Medical therapy for coronary artery disease
works, even (especially) in the real world. Am J Med 110:497498
5. MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J et al
(1990) Blood pressure, stroke, and coronary heart disease. Part 1,
prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet
335:765774
6. Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Eberlein
KA et al (1990) Blood pressure, stroke and coronary heart disease. Part 2, short term reductions in blood pressure: overview of

Intern Emerg Med (2011) 6 (Suppl 1):S45S51

7.

8.

9.

10.
11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

randomised drug trials in their epidemiological context. Lancet

335:827838
Andersson OK, Almgren T, Persson B, Samuelsson O, Hedner T,
Wilhelmsen L (1998) Survival in treated hypertension: follow up
study after two decades. BMJ. 317:167171
Glynn RJ, LItalien GJ, Sesso HD, Jackson EA, Buring JE (2002)
Development of predictive models for long-term cardiovascular
risk associated with systolic and diastolic blood pressure.
Hypertension. 39:105110
Blacher J, Evans A, Arveiler D, Amouyel P, Ferrie`res J, Bingham
A, Yarnell J, Haas B, Montaye M, Ruidavets J-B, Ducimetie`re P,
PRIME Study Group (2010) Residual cardiovascular risk in
treated hypertension and hyperlipidaemia: the PRIME Study.
J Human Hypertension 24:1926
Libby PL (2005) The forgotten majority. Unfinished business in
cardiovascular risk reduction. J. Am. Coll. Cardiol 46:12251228
Fruchart J-C, Sacks F, Hermans MP et al (2008) Executive
statement: the Residual Risk Reduction Initiative: a call to action
to reduce residual vascular risk in dyslipidemic patients. A condensed position paper by the Residual Risk Reduction Initiative
(R3i). Diab. Vasc. Dis. Res 5:319335
Grundy SM, Cleeman JI, Daniels SR et al (2005) American Heart
Association; National Heart, Lung, and Blood Institute: Diagnosis and management of the metabolic syndrome: an American
Heart Association/National Heart, Lung, and Blood Institute
Scientific Statement. Circulation 112:27352752
Graham I, Atar D, Borch-Johnsen K et al (2007) European
Society of Cardiology (ESC) Committee for Practice Guidelines
(CPG): European guidelines on cardiovascular disease prevention
in clinical practice. Fourth Joint Task Force of the European
Society of Cardiology and other Societies on Cardiovascular
Disease Prevention in Clinical Practice. Executive summary. Eur.
Heart J. 28:23752414
Ryden L, Standl E, Bartnik M et al (2007) Guidelines on diabetes,
prediabetes, and cardiovascular diseases: executive summary.
The Task Force on Diabetes and Cardiovascular Diseases of the
European Society of Cardiology (ESC) and of the European
Association for the Study of Diabetes (EASD). Eur. Heart J.
28:88136
Brown G, Albers JJ, Fisher LD et al (1990) Regression of coronary artery disease as a result of intensive lipid-lowering therapy
in men with high levels of apolipoprotein B. N Engl J Med
323:12891298
HPS Collaborative Group: MRC/BHF Heart Protection Study of
Cholesterol-Lowering with Simvastatin in 5, 963 People with
Diabetes: a randomised placebo-controlled trial. (2003) Lancet
361: 20052016
Sacks FM, Tonkin AM, Shepherd J et al (2000) Effect of pravastatin on coronary disease events in subgroups defined by coronary risk factors: the Prospective Pravastatin Pooling Project.
Circulation 102:18931900
Colhoun HM, Betteridge DJ, Durrington PN et al (2004) CARDS
investigators: Primary prevention of cardiovascular disease with
atorvastatin in Type 2 diabetes in the Collaborative Atorvastatin
Diabetes Study (CARDS): multicentre randomised placebo-controlled trial. Lancet 364:685696
LaRosa JC, Grundy SM, Waters DD et al (2005) Intensive lipid
lowering with atorvastatin in patients with stable coronary disease. N Engl J Med 352:14251435
Shepherd J, Barter P, Carmena R et al (2006) Treating to New
Targets Investigators: Effect of lowering LDL cholesterol substantially below currently recommended levels in patients with
coronary heart disease and diabetes: The Treating to New Targets
(TNT) study. Diabetes Care 29:12201226
Ridker PM, Danielson E, Fonseca FA, JUPITER Study Group
et al (2008) Rosuvastatin to prevent vascular events in men and

S51

22.

23.

24.

25.

26.

27.

28.

29.
30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

women with elevated C-reative protein. N. Engl. J. Med

359:21952207
Taylor AJ, Villines TC, Stanek EJ et al (2009) Extended-Release
Niacin or Ezetimibe and Carotid Intima-Thickness. New Engl J
Med 361:21132122
Ginsberg HN, Elam MB, Lovato LC, The ACCORD Study Group
et al (2010) Effects of combination lipid therapy in type 2 diabetes. New Engl J Med 362:15631574
Carey VJ, Bishop L, Laranjo N et al (2010) Contribution of high
plasma triglyceride and low HDL-C to residual risk or coronary
heart disease after establishment of LDL-C control. Am J Cardiol
106:757763
Birjmohun RS, Hutten BA, Kastelein JJ et al (2005) Efficacy and
safety of HDL-C increasing compounds. J Am Coll Cardiol
45:185197
Ridker PM, Genest J, Boekholdt AM et al (2010) HDL-C and
residual risk of first cardiovascular events after treatment with
potent statin therapy: an analysis of JUPITER trial. Lancet
376:333339
Barter P, Gotto AM, LaRosa JC et al (2007) HDL-C, very low
levels of LDL-C and cardiovascular events. N Engl J Med
357:13011310
Cannon CP, Braunwald E, McCabe CH et al (2004) Intensive
versus moderate lipid lowering with statins after acute coronary
syndromes. N Engl J Med 350:14951505
Nicholls SJ, Tuzcu EM, Sipahi I et al (2007) Statins, HDL-C and
regression of coronary atherosclerosis. JAMA 297:499508
Cui Y, Watson DJ, Girman CJ et al (2009) Effects of increasing
HDL-C and decreasing LDL-C on the incidence of first acute
coronary event from the Air Force/Texas Coronary Atherosclerosis Prevention Study. Am J Cardiol 104:829834
Downs J, Clearfield M, Weis S et al (1998) Air Force/Texas
Coronary Atherosclerosis Prevention study. Primary prevention
of acute coronary events with lovastatin in men and women with
average cholesterol levels. JAMA 279:16151622
Lloyd-Jones DM, Dyer AR, Wang R, Daviglus ML, Greenland P
(2007) Risk factor burden in middle age and lifetime risks for cardiovascular and non-cardiovascular death. (Chicago Heart Association Detection Project in Industry). Am J Cardiol 99:535540
Cohen JC, Boerwinkle E, Mosely TH, Hobbs HH (2006)
Sequence variations in PCSK9, low LDL and protection against
coronary heart disease. N Engl J Med 354:1264
Cholesterol Treatment Trialists Collaboration. Efficacy and
safety of cholesterol lowering treatment: prospective meta-analysis of data from 90056 participants from 14 randomized trials of
statins. (2005) Lancet 366: 1267-78
Pedersen TR, Cater NB, Faergeman O et al (2010) Comparison of
atorvastatin 80 mg/day versus simvastatin 20 to 40 mg/day on
frequency of cardiovascular events late (five years) after acute
myocardial infarction (from the Incremental in End Points
through Aggressive Lipid Lowering (IDEAL) Trial. Am J Cardiol
106:354359
Ford I, Murray H, Packard C et al (2007) Long term follow-up of
the West of Scotland Coronary Prevention Study. N Engl J Med
357:14771486
Kones R (2011) Is prevention a fantasy, or the future of medicine? A panoramic view of recent data, status, and direction in
cardiovascular prevention. Ther Adv Cardiovasc Dis. 5:6181
Ferrario M, Chiodini P, Chambless LE et al (2005) CUORE
Project Research Group. Prediction of coronary events in a low
incidence population. Assessing accuracy of the CUORE cohort
study prediction equation. Int J Epidemiol 34:413421
WHO MONICA Project. MONICA manual (19981999) Geneva.
Switzerland: World Health Organization, Office of Cardiovasculr
Diseases. Available from: URL: http://www.ktl.fi/publications/
monica/manual/index.htm

123