CARDIOVASCULAR JOURNAL OF AFRICA • Vol 22, No 1, January/February 2011
AFRICA
51
ASCOT analysis with atorvastatin shows limits of CRP as
indicator of cardiovascular risk
The addition of high-sensitivity C-reactive
protein (hs-CRP) measurements did not
much improve conventional risk assess-
ments in patients with hypertension and
other cardiovascular (CV) risk factors
in a
post hoc
analysis of an Anglo–
Scandinavian Cardiac Outcomes Trial
(ASCOT) lipid-lowering arm.
1
The lipid-lowering arm of ASCOT,
which involved more than 10 000 patients,
showed that atorvastatin therapy achieved
significant reduction in major cardiovas-
cular events in hypertensive patients with
normal or modestly elevated low-density
lipoprotein cholesterol (LDL-C) levels.
Also, the combination of atorvastatin/
amlodipine as analysed in both the blood
pressure-lowering and lipid arms of the
ASCOT trial showed a greater reduc-
tion in non-fatal myocardial infarction
(MI) and fatal coronary artery disease
compared to patients receiving the ateno-
lol-based therapy.
This
post hoc
ASCOT assessment
of the value of hs-CRP was based on
data from 485 patients from the UK and
Irish centres who reached a composite
endpoint that included CV death, non-
fatal MI, coronary revascularisation, or
fatal/non-fatal stroke, and who were age
and gender matched to 1 367 patients in
the control group who hadn’t suffered
one of those events. They were followed
for 5.5 years.
Baseline levels of hs-CRP and LDL-C
were significantly correlated with each
other at
p
<
0.0001, and both parameters
were significantly predictive of a compos-
ite CV endpoint that included CV death,
non-fatal MI, coronary revascularisation,
or fatal/non-fatal stroke, at an odds ratio
(OR) of 1.31 (95% CI: 1.10–1.56,
p
=
0.002) for LDL-C and OR 1.19 (95% CI:
1.05–1.34,
p
=
0.006) for every one stand-
ard deviation increase in hs-CRP.
According to Dr Peter Sever, Imperial
College, London, including hs-CRP
into modified and full Framingham risk
models only minimally improved predic-
tion of cardiovascular events, while the
addition of on-treatment hs-CRP meas-
urement to on-treatment LDL-C measure-
ment did not improve the prediction of
patient response to the statin therapy.
Also, among patients assigned to ator-
vastatin:
•
Baseline hs-CRP did not predict the
magnitude of the drug’s effect on
cardiovascular endpoints.
•
Over six months of atorvastatin ther-
apy, the median LDL-C fell by 40.3%
and the median hs-CRP dropped by
27.4%.
Lower on-treatment LDL-C but not
hs-CRP was associated with a highly
significant reduction in cardiovascular
events over six months:
•
An LDL-C less than the median of
2.1 mmol/l, compared with LDL-C at
or higher than the median predicted a
reduction in the composite endpoint
at an adjusted OR of 0.41 (95% CI:
0.22–0.75,
p
=
0.004).
•
But an hs-CRP less than the median
of 1.83 mg/l, compared to hs-CRP at
or higher than the median, showed no
such predictive effect at an adjusted
OR of 0.86 (95% CI: 0.49–1.51,
p
=
0.60).
‘It is clear that achievement of low LDL
cholesterol levels confers great benefit,
and there’s no added benefit by achieving
a lower CRP level’, Sever said in his pres-
entation. Among patients in the analysis
who had been randomised to receive ator-
vastatin, a significant reduction in LDL-C
corresponded to a significant drop in CV
event risk at six months. But a significant
fall in hs-CRP levels did not predict a
decrease in CV events.
‘These results do not support current
proposals to measure CRP in the clinical
setting, either to assign statins to individu-
als on the basis of an elevated CRP alone
or to monitor CRP levels as an indicator
of the efficacy of statin treatment’, he
said.
The ASCOT analysis, according to
Sever, challenges the recent broadening
of rosuvastatin indications by the US
FDA, based on the JUPITER trial, which
compared the drug with placebo in more
than 17 000 adults with normal LDL-C
and hs-CRP
≥
2.0 mg/l.
2
Sever also pointed out that in the
ASCOT analysis, adding CRP to a conven-
tional risk assessment based on elements
of the Framingham risk score improved
risk prediction, but not by much. ‘If you
measure their total cholesterol-to-HDL
cholesterol ratio, if you measure their
blood pressure, if you’ve decided whether
or not they smoke, [and consider] their
age and their sex, that tells you all you
need to know about their risk’, he said.
‘Putting CRP in there doesn’t alter the
[risk-status] classification of people, and
that suggests to me that it doesn’t have
added value.’
That means CRP does not have a
practical role in deciding whether to initi-
ate statin therapy, according to Sever. ‘In
patients with hypertension, they’re on
statins. In patients with diabetes, they’re
on statins. In patients with stable coronary
artery disease, they’re on statins. There’s
no additional benefit to be gained from
measuring CRP.’
Discussing the ASCOT analysis, Dr
Elliot Antman (Brigham and Women’s
Hospital) seemed to agree that CRP prob-
ably has a limited risk-stratification role.
‘For patients with a very low risk-factor
profile [based on standard risk assess-
ment], a measurement of CRP might be
considered as a sort of general screen’,
he said. ‘But I’m not so convinced that
it’s going to provide a lot of incremental
value for deciding who should start on a
statin.’
‘If all one has to go on is CRP level,
there’s evidence that it can be predic-
tive’, Antman observed. ‘But if you start
giving me incremental information like
this person smokes, or has diabetes, or has
hypertension, then the amount of benefit
I’m gaining from knowing the person’s
CRP is dropping with each new piece of
information you give me.’
1. Amended from a Heartwire report. American
Heart Association 2010.
2. Sever P. The Anglo-Scandinavian Cardiac
Outcomes Trial (ASCOT): testing C-reactive
protein at baseline and on-treatment as
an independent predictor of cardiovascu-
lar outcomes. American Heart Association
2010 scientific sessions. November 17, 2010;
Chicago, IL. Late-breaking clinical trials IV.
3. Ridker PM, Danielson E, Fonseca FA,
et al
.
Rosuvastatin to prevent vascular events in
men and women with elevated C-reactive
protein.
New Engl J Med
2008;
359
: 2195–
2207.
