 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
07-November-2003, Vol 116 No 1185
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Secondary prevention in coronary artery disease patients in
South Auckland: moving targets and the current treatment gap
Seifeddin El-Jack and Andrew Kerr
A wealth of clinical-trial and epidemiological data supports
the utility of lifestyle modification, smoking cessation, anti-hypertensive and
lipid-lowering therapy, anti-platelet therapy and beta blockade in secondary
prevention of coronary artery disease
(CAD).1–4 Disappointingly, international
and local studies have demonstrated a significant disparity between ideal
secondary prevention and the clinical
reality.5–12 The reasons for this
‘treatment gap’ are multifactorial. The 1996 New Zealand Heart
Foundation (NZHF) Guidelines,13 current at the
time of this audit, recommended that lipid-lowering strategies are instigated in
patients with proven CAD and total cholesterol (TC) ≥5.5 mmol/l, aiming
for target TC of 3–5 mmol/l. In December 1998 the New Zealand
Government’s pharmaceutical watchdog, PHARMAC, adopted this level as a
cut-off point for funding statins. Following revascularisation (coronary artery
bypass grafting (CABG) or percutaneous coronary intervention), the cut-off point
for funding therapy was TC ≥4.5 mmol/l. More recently, the European
Societies Guidelines proposed a target low-density lipoprotein (LDL) <3
mmol/l,14 and the National Cholesterol
Education Programme Adult Treatment Panels II15
and III16 (NCEP-ATPII and III) a target LDL
<2.6 mmol/l (100 mg/dl).
From early 2002 PHARMAC no longer required a special
authority for prescription of the statin simvastatin, making it more widely
available. In mid 2002 the landmark Heart Protection Study was published, which
provided evidence for benefit from statin use in patients with CAD regardless of
baseline cholesterol levels.17 We are now in an
environment where we have a substantial treatment gap, resulting from a change
in the evidence base, together with the opportunity to close that gap due to a
change in PHARMAC policy.
In 2001 we audited lipid, blood pressure (BP) status,
anti-platelet therapy and smoking status in patients with prior CABG or recent
myocardial infarction (troponin-positive acute coronary syndrome). The aims of
this audit were to:
We also have the opportunity to estimate how many
additional patients should now be considered for statin therapy because of the
Heart Protection Study results and the change in prescribing limitations for
statins.
MethodsData were collected between May
and July 2001 for the retrospective study of three groups of patients. The first
group were consecutive patients who had myocardial infarction (MI) in 1999 who
did not subsequently have coronary intervention, and therefore required a total
cholesterol ≥5.5 mmol/l to receive a funded statin (recent MI group). The
choice of this temporal cohort was to allow time for risk-factor modification to
have been achieved at the time of data collection. The second group were
consecutive patients who had CABG surgery in the year 1999 (recent CABG). The
third group were consecutive patients who had CABG in the year 1994 (remote
CABG). The latter were chosen as at that time the recommendations of the 4S
trial were not yet fully
implemented.1
The recent MI patients were identified from Middlemore
Hospital Coronary Care Unit records, and patients in the other two groups from
the Green Lane Hospital Cardiothoracic Department registry. Data were obtained
from patient records and by contacting general practitioners.
Consecutive patients were screened to obtain 50
patients in each group. Patients were excluded for the following reasons: death
(n = 13), and loss to follow up or significantly incomplete records (n = 30).
After entry into the audit, two patients in the recent MI group were found to be
deceased and one patient in the remote CABG group moved overseas. They were
excluded from analysis. Therefore, 147 patients were audited.
For each patient the following data were obtained:
updated lipid profiles (within 12 months), last recorded blood pressure,
anti-platelet therapy, and smoking status. In addition, the lipid profiles at
the time of the index event (MI or CABG) were obtained together with HbA1c in
known diabetics, and current use of angiotensin-converting enzyme inhibitors
(ACEI) and/or beta-blocker therapy. Left ventricular (LV) function was noted (LV
ejection fraction (LVEF) <50% was considered impaired).
Descriptive data are expressed as mean ± standard
deviation. Samples were compared using chi-square testing. Differences were
considered statistically significant for p <0.05.
ResultsTable 1. Patient characteristics at baseline
Lipid data Baseline
lipid profiles were available in 139 patients (95%). Fasting levels were
available in only 42 out of 147 patients (29%). Current lipid data were
available in all patients. Of these, 113 had levels documented as fasting
results. Of the remaining 34 patients, 28 were on statins of whom 16 had an LDL
<2.6 mmol/l. Of those not on a statin the mean LDL (± SD) was 2.5
(± 0.2) mmol/l, which is lower than for patients with fasting data (2.8
± 0.9 mmol/l). Thus, the failure to obtain fasting data in these patients
did not appear to systematically bias the results in the direction of poorer
control. For the purpose of this audit, TC and LDL data were analysed as a mix
of fasting and non-fasting levels. Only fasting triglyceride levels are
reported.
Table 2 shows the group and overall lipid data at baseline
and currently. There is no significant difference in the current lipid fractions
between any of the groups. Twenty nine per cent of our study population were not
on a statin medication. Seventeen patients (12% of the whole group) not on a
statin did not meet the contemporaneous PHARMAC cut-off point. Twenty per cent
of those whose TC exceeded the PHARMAC cut-off point were not on therapy. Forty
per cent of those with LDL over the NCEP target of 2.6 mmol/l and 36% of those
over the contemporaneous NZHF target TC of 3–5mmol/l were not on a statin
(Table 3). Conversely, there are many patients who do not meet target lipid
levels despite being on a statin. This is dependent on the target set. For
example, of 77 patients (55%) who do not meet the NCEP target LDL of <2.6
mmol/l, 46 (60%) are already on a statin.
Table 2. Mean lipid fractions (mmol/l) both at baseline
and currently, and current use of lipid-lowering therapy
TC = total cholesterol; LDL = low-density lipoprotein;
HDL = high-density lipoprotein; TG = triglycerides
Table 3. The percentage of patients not achieving
defined guideline targets (mmol/l) and their use of statins
TC = total cholesterol; LDL = low-density lipoprotein;
HDL = high-density lipoprotein; TG = triglycerides
Attainment of lipid targets
(Table 4) As expected, the more rigorous targets are met by fewer
patients. Whilst 69% met the European Cardiac Society LDL target of <3.0
mmol/L, only 45% meet the more rigorous NCEP target of <2.6 mmol/l and only
13% meet the TC target of <4 mmol/l from the NZHF 2002 Interim Consensus
Statement for the Management of
Dyslipidaemia.18
Table 4. Patients meeting risk-factor targets
TC = total cholesterol; LDL = low-density lipoprotein;
HDL = high-density lipoprotein; TG =
triglycerides
NB: lipid targets in mmol/l; fasting TG levels available in 111 patients CABG patients were more likely to attain a TC ≤5
mmol/l and LDL <2.6 mmol/l than those who had MI without revascularisation (p
= 0.052 and 0.005 for TC and LDL respectively). The overall use of statins,
however, was not significantly different between the two groups (p =
0.374)
Other secondary prevention
factors Eighty eight per cent of the overall group were currently
non-smokers; however, 23% of the recent MI group were continuing to smoke. This
was a higher rate than for CABG patients (p = 0.014).
Mean BP (± SD) for the whole group was 135/78 (±
16/9) mmHg. The target BP of 140/90 was not met in 41% of patients. There was no
significant difference between CABG and MI patients (p = 0.100).
Data on HbA1c were available in 28 of 35 diabetic patients
(80%). The mean HbA1c (± SD) was 7.5% (± 1.5) and 61% of patients had
a sub-optimal HbA1c level above 7%.
Data regarding anti-platelet and ACEI usage were available
in all patients and in 146 patients for beta blockers. Ninety three per cent
were on aspirin (of the 10 patients not on aspirin, one was on warfarin and
another had experienced a previous haemorrhagic stroke), 34% on an ACEI and 45%
on a beta blocker. Beta blockers were used more frequently in MI patients than
those who had CABG (p = 0.008). There was no significant difference in the use
of ACEI (p = 0.267). In patients with LV impairment (23 out of 104 patients with
documented LV function) only 70% were
on ACEI and 43% on beta blockers.
All targets met The
AHA/ACC set secondary prevention targets, which include LDL <2.6 mmol/l, BP
<140/90 and a mandatory anti-platelet agent and
non-smoking.19 Of the 140 patients for whom all
these data were available, only 31 (22%) met all targets. This analysis excludes
other possible targets including other lipid fractions, HbA1c, ACEI and
beta-blocker use, and weight reduction, which if included would reduce this
figure further. Using the contemporaneous NZHF target TC of 3–5 mmol/l,
only 44 patients (30%) met all targets.
DiscussionIn this group of patients with
known coronary artery disease, 55% had an LDL cholesterol ≥2.6 mmol/l, 37%
a TC >5 mmol/l, 41% had BP >140/90 mmHg, 12% still smoked and 7% were not
on anti-platelet therapy. Only 30% of patients (44/147) met all secondary
prevention targets at the time the audit was conducted. There is a formidable
gap between ideal management of risk factors for cardiovascular disease and the
clinical reality. The reasons for this treatment gap are clearly multifactorial
and include patient, doctor, cultural and organisational/government factors.
Similar results were reported in recent European and North American studies
looking at secondary prevention
profiles.7–12
A relatively high percentage (71%) of the whole group were
on a statin. Thirteen patients from the recent MI group (27%) whose total
cholesterol levels exceeded the PHARMAC cut-off point are now eligible for
treatment on the NZHF 2002 Interim Consensus Statement for the Management of
Dyslipidaemia. The recent Heart Protection Study results suggest that all
patients included in this audit would benefit from statin treatment. With the
recent liberalisation of statin availability in NZ this is an important gap that
can be closed. Of concern, we found that some patients were not on a statin
despite appearing to meet PHARMAC cut-off points. A further problem is the
number of patients who do not achieve target lipid profiles despite being on a
statin, suggesting that they may benefit from more aggressive pharmacological
and lifestyle intervention.
Only 34% of patients were taking an ACEI. The recent HOPE
study suggests that most if not all these patients would benefit from an
ACEI.20 This and the modest use of beta
blockers are other important treatment gaps to be closed.
These results may not be representative of our general
population with CAD, as two thirds of our study group were CABG patients and
therefore statin use and other secondary prevention could be better than in the
non-intervention population. Although not all reaching statistical significance,
the data suggest this. For instance, 82% of the recent CABG patients were on
statins as opposed to 67% of the recent MI patients. Furthermore, 87% of
patients in the recent CABG group with a TC ≥5mmol/l were on statins
compared with 54% of the patients in the recent MI group with the same
TC.
Also risk-factor modification seems to be more effective in
the CABG groups, with a TC ≤5 mmol/l achieved in 70% compared with 53% in
the non-interventional group (p = 0.052). An LDL <2.6 mmol/l was achieved in
54% of CABG patients but only 27% of the recent MI patients (p = 0.005). This
may result from the more lenient PHARMAC statin eligibility criteria for CABG
patients prevailing at that time, but other unmeasured factors such as
differential compliance with lifestyle management may explain the difference.
This is suggested by the finding that smoking was less common in CABG patients
than those with MI (93% versus 79%, p = 0.014).
We found that the risk-factor management of patients who had
their CABG operations in 1994 was similar to more recent patients. These
patients were operated on before the 4S trial, which made statin use mandatory
in many of them.1 This may partly reflect
survivor bias but suggests that these patients’ GPs are modifying
treatment appropriately as evidence changes.
This study also illustrates an important problem in
measuring the treatment gap: deciding on appropriate targets. It is relatively
easy to set a target for binary risk factors such as smoking; it is much harder
to set a target for lipid levels or blood pressure where there are few data to
support a particular target level. For instance in the Heart Protection Study,
patients with baseline LDL cholesterol of <2.6 mmol/l, the NCEP target,
benefited from addition of a statin. This leads to inconsistency in target
setting with major implications when the treatment gap is being assessed. For
example, in this audit 63% reached the 1996 NZHF target TC of ≤5.0 mmol/l,
69% reached the European Cardiac Society target LDL of <3.0 mmol/l, 45%
reached the NCEP target of <2.6 mmol/l and only 13% reached the recent NZ
consensus target TC of <4.0 mmol/l.
Improved strategies to improve risk-factor management in
these patients are needed. These include using information technology to
disseminate patient-specific, guideline-based treatment recommendations across
primary and secondary care, and specifically funding GPs to more actively manage
patients whose risk factors are sub-optimally controlled.
In this retrospective analysis some patients were not
included due to death, incomplete records or other loss to follow up. This was
most frequent in the early CABG group and it is likely that the treatment gap is
even greater in those lost to follow up. The resulting survivor bias may have
led to an overestimation of the adequacy of secondary prevention in this group.
The study is underpowered to conclusively show a difference in secondary
prevention between early and late CABG groups, and the MI and CABG groups,
respectively.
In summary, risk-factor management is sub-optimal in a
significant proportion of secondary prevention patients. The treatment gap may
be greater in those with CAD who have not had a cardiac intervention. Improved
strategies to improve risk-factor management in these patients are
needed.
Author information:
Seifeddin El-Jack, Cardiology Registrar; Andrew Kerr, Cardiologist, Cardiology
Department, Middlemore Hospital, Auckland
Acknowledgements:
This audit was funded by a research grant from MSD New Zealand and the
Cardiology Research Fund at Middlemore Hospital.
Correspondence: Dr
Andrew Kerr , Middlemore Hospital, P O Box 93311, Otahuhu, South
Auckland. Fax: (09)
250
0259;
andrewkerr@middlemore.co.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|